RU2798759C2 - Parametric encoding and decoding of multi-channel audio signals - Google Patents

Abstract

FIELD: computer technology.

SUBSTANCE: invention relates to the coding of audio signals. An audio decoding method, including reception of a two-channel downmix signal and upmix parameters for restoration of a M-channel audio signal based on the downmix signal, signal reception (S) indicating the selected one of at least two encoding formats of the M-channel audio signal divided into respective first and second groups of one or more channels, calculating the first upmix signal as a linear display of the downmix signal, calculating the second upmix signal as a linear display of the downmix signal, and combining the first and second signals upmixing to obtain a multidimensional reconstructed signal corresponding to the M-channel sound signal, while M-channel sound signal has a predetermined channel configuration, and said selected coding format is switched between at least two coding formats.

EFFECT: increased reliability when decoding the audio signal.

18 cl, 17 dwg, 1 tbl

Description

Cross references to related applications

This application claims priority over U.S. Provisional Application No. 62/073,642, filed October 31, 2014, and U.S. Provisional Application No. 62/128,425, filed March 4, 2015, the entire contents of which are hereby incorporated herein by links.

Technical field

The invention described in this application essentially relates to parametric coding and decoding of audio signals and, in particular, to parametric coding and decoding of channelized audio signals.

Prerequisites for the creation of the invention

In order to reproduce audio data represented by a multi-channel audio signal, audio reproduction systems comprising a plurality of speakers are often used, with the respective channels of the multi-channel audio signal reproduced through the respective speakers. The multi-channel audio signal may, for example, be recorded with a plurality of acoustic transducers, or may be created with audio production equipment. In many cases, there are bandwidth limitations when transmitting the audio signal to the playback equipment, and/or there is limited space in the computer storage device or portable storage device for storing the audio signal. Audio coding systems exist for parametrically encoding audio signals to reduce bandwidth or data storage size. At the encoder side, these systems typically downmix the multi-channel audio signal to a downmix signal, which is typically a mono (single-channel) or stereo (two-channel) downmix signal, and extract additional information describing the characteristics of the channels through parameters such as differences levels and cross-correlation. The downmix signal and side information are then encoded and transmitted to the decoder side. On the decoder side, based on the downmix signal, taking into account the parameters of additional information, reconstruction, i.e., approximation, of the multichannel audio signal is performed.

In view of the variety of different types of devices and systems suitable for the reproduction of multichannel audio data, including tools in the recently emerging segment focused on home use by end users, it is urgent to create new and alternative ways to efficiently encode multichannel audio data that would reduce bandwidth requirements. bandwidth and/or storage capacity, make it more convenient to reconstruct the multi-channel audio signal on the decoder side, and/or increase the fidelity of the reproduction of the multi-channel audio signal reconstructed on the decoder side.

Brief description of the drawings

The embodiments will be described in more detail below and with reference to the accompanying drawings, wherein:

in FIG. 1 and 2 are generalized block diagrams of coding sections for encoding M -channel audio signals as two-channel downmix signals and associated upmix parameters in accordance with exemplary embodiments;

in FIG. 3 is a generalized block diagram of an audio coding system comprising the coding section shown in FIG. 1, in accordance with the exemplary implementation;

in FIG. 4 and 5 are flowcharts of audio coding methods for encoding M -channel audio signals as two-channel downmix signals and associated upmix parameters in accordance with exemplary embodiments;

in FIG. 6-8 show alternative methods for dividing 11.1-channel (or 7.1 + 4-channel or 7.1.4-channel) audio signal into channel groups represented by respective downmix channels, in accordance with exemplary embodiments;

in FIG. 9 is a generalized block diagram of a decoding section for reconstructing an M -channel audio signal based on a two-channel downmix signal and associated upmix parameters in accordance with an embodiment;

in FIG. 10 is a generalized block diagram of an audio decoding system comprising the decoding section shown in FIG. 9, in accordance with the exemplary implementation;

in FIG. 11 is a generalized block diagram of the mixing section included in the decoding section shown in FIG. 9, in accordance with the exemplary implementation;

in FIG. 12 is a flowchart of an audio decoding method for reconstructing an M -channel audio signal based on a two-channel downmix signal and associated upmix parameters, in accordance with an embodiment; And

in FIG. 13 is a generalized block diagram of a decoding section for reconstructing a 13.1-channel audio signal based on a 5.1-channel signal and associated upmix parameters in accordance with an embodiment;

in FIG. 14 is a generalized block diagram of an encoding section configured to determine a suitable encoding format that can be used to encode an M -channel audio signal (and possible subsequent channels), and represent the M -channel audio signal as a two-channel downmix signal and associated upmix parameters for selected format;

in FIG. 15 details the dual mode downmix section in the coding section shown in FIG. 14;

in FIG. 16 details the dual-mode analysis section in the encoding section shown in FIG. 14; And

in FIG. 17 is a flowchart of an audio coding method that may be implemented by the components shown in FIG. 14–16.

All figures are schematic and essentially show those parts that are necessary to explain the essence of the invention, while other parts may be omitted or simply implied.

Description of exemplary embodiments

In the present application, an audio signal may be a single audio signal, an audio portion of an audiovisual signal or a multimedia signal, or any of such signals combined with metadata. In the present application, a channel is an audio signal associated with a predetermined/fixed spatial position/orientation or an undefined spatial position such as "left" or "right".

I. Overview. Decoder side

According to the first aspect, the embodiments provide audio decoding systems, audio decoding methods, and related computer program products. The proposed decoding systems, methods, and computer program products according to the first aspect may substantially share the same general features and benefits.

According to exemplary embodiments, an audio decoding method is provided that includes obtaining a two-channel downmix signal and upmix parameters for parametrically reconstructing an M -channel audio signal based on the downmix signal, wherein M ≥ 4. The audio decoding method includes obtaining a signaling indicative of a selected one of at least two coding formats of the M -channel audio signal, the coding formats being mapped to respective different channelizations of the M -channel audio signal into respective first and second groups of one or more channels. In this coding format, the first channel of the downmix signal corresponds to a linear combination of the first group of one or more channels of the M -channel audio signal, and the second channel of the downmix signal corresponds to the linear combination of the second group of one or more channels of the M -channel audio signal. The audio decoding method further includes determining a set of pre-decorrelation coefficients based on the specified coding format; calculating the decorrelation input signal as a linear display of the downmix signal, wherein a set of pre-decorrelation coefficients is applied to the downmix signal; generating a decorrelated signal based on the input decorrelation signal; determining sets of upmix coefficients of the first type, which are called "wet" upmix coefficients in the present application, and the second type, which are called "dry" upmix coefficients in the present application, based on the obtained upmix parameters and the specified coding format; calculating a first type upmix signal, which is referred to as a "dry" upmix signal in the present application, as a linear display of the downmix signal, wherein a set of "dry" upmix coefficients is applied to the downmix signal; calculating a second type of upmix signal, which is referred to in the present application as a "wet" upmix signal, as a linear display of the decorrelated signal, wherein a set of "wet" upmix coefficients is applied to the decorrelated signal; and combining the dry and wet upmix signals to obtain a multi-dimensional reconstructed signal corresponding to the M -channel audio signal to be reconstructed.

Depending on the audio data of the M -channel audio signal, different divisions of the channels of the M -channel audio signal into first and second groups, with each group contributing to the channel of the downmix signal, may be suitable, for example, to facilitate the reconstruction of the M -channel audio signal from the downmix signal. mixing, thereby increasing the (perceived) fidelity of the M -channel audio signal reconstructed from the downmix signal and/or increasing the coding efficiency of the downmix signal. That the audio decoding method is capable of obtaining signaling indicative of the selected coding format and adapting the definition of the pre-decoration coefficients and the wet and dry upmix coefficients for the specified coding format allows the selection of the coding format at the encoder side, for example, based on the audio data of the M -channel audio signal to take advantage of the comparative advantages of using this particular coding format to represent the M -channel audio signal.

In particular, determining the pre-decorrelation coefficients based on the specified coding format may select the channel or channels of the downmix signal from which the decorrelated signal is generated and/or determine the significance of such channel or channels based on the specified coding format prior to generating the decorrelated signal. Thus, the fact that the audio decoding method allows the pre-decoration coefficients to be determined differently for different coding formats can increase the fidelity of the reconstructed M -channel audio signal.

The first channel of the downmix signal may, for example, be formed on the encoder side as a linear combination of the first group of one or more channels in accordance with a specified coding format. Similarly, the second channel of the downmix signal can, for example, be formed on the encoder side as a linear combination of a second group of one or more channels in accordance with a specified coding format.

The channels of an M -channel audio signal may form, for example, a subset of a larger number of channels, which together represent a sound field.

The decorrelated signal serves to increase the dimension of the audio data of the downmix signal as perceived by the listener. Creating a decorrelated signal may include, for example, applying a linear filter to the decorrelation input signal.

By computing the decorrelation input signal as a linear mapping of the downmix signal, it is meant that the decorrelation input signal is obtained by first linearly transforming the downmix signal. The first linear transform takes two channels of the downmix signal as an input and provides the channels of the input decorrelation signal as an output, and the pre-decrelation coefficients are coefficients quantifying this first linear transform.

By calculating the dry upmix signal as a linear mapping of the downmix signal, it is meant that the dry upmix signal is obtained by a second linear transformation of the downmix signal. The second linear conversion takes two channels of the downmix signal as an input and provides M channels as an output, and the dry upmix coefficients are coefficients quantifying this second linear conversion.

By computing the wet upmix signal as a linear mapping of the decorrelated signal, it is meant that the wet upmix signal is obtained by a third linear transformation of the decorrelated signal. The third linear transform takes the channels of the decorrelated signal as an input and provides M channels as an output, and the wet upmix coefficients are coefficients that quantify this third linear transform.

Combining the dry and wet upmix signals may include adding audio data from the respective channels of the dry upmix signal to the audio data of the respective channels of the wet upmix signal, for example, by using additive mixing by samples or by transform coefficients.

The signaling may be obtained, for example, together with the downmix signal and/or the upmix parameters. The downmix signal, upmix parameters, and signaling may, for example, be extracted from the bitstream.

In an exemplary embodiment, the condition M = 5 may be present, i.e., the M -channel audio signal may be a five-channel audio signal. The audio decoding method according to the present embodiment can be used, for example, to reconstruct five standard channels in one of the currently established 5.1 audio formats from a two-channel downmix signal of these five channels, or to reconstruct five channels on the left side or right side into 11.1 multi-channel the audio signal from the two-channel downmix signal of these five channels. In an alternative embodiment, the condition M = 4 or M ≥ 6 may be present.

In an exemplary embodiment, each of the decorrelation input and the decorrelated signal may contain M −2 channels. In the present embodiment, a decorrelated signal channel may be created based on at most one decorrelation input signal channel. For example, each decorrelated signal channel can be created based on at most one decorrelation input channel, but different decorrelated signal channels, for example, can be created based on different decorrelation input channels.

In the present exemplary embodiment, the pre-decrelation coefficients may be determined such that, in each of the coding formats, a decorrelation input signal channel receives a contribution from at most one downmix signal channel. For example, the pre-decrelation coefficients may be determined such that, in each of the coding formats, each channel of the decorrelation input signal matches a channel of the downmix signal. However, it should be understood that at least some of the channels of the decorrelation input signal may coincide with, for example, different channels of the downmix signal in a particular coding format and/or in different coding formats.

Since, in a particular coding format, the two channels of the downmix signal represent separate first and second groups of one or more channels, the first group may be reconstructed from the first channel of the downmix signal using, for example, one or more channels of a decorrelated signal generated from the first channel of the signal. downmix signal, and the second group may be reconstructed from the second channel of the downmix signal using, for example, one or more channels of the decorrelated signal generated from the second channel of the downmix signal. In the present embodiment, each coding format may not allow contributions from the second group of one or more channels into a reconstructed version of the first group of one or more channels by means of a decorrelated signal. Likewise, in each coding format, contributions from a first group of one or more channels to a reconstructed version of a second group of one or more channels via a decorrelated signal may be prevented. Thus, in the present embodiment, an increase in the reproduction fidelity of the reconstructed M -channel audio signal can be proposed.

In an exemplary embodiment, the pre-decrelation coefficients may be determined such that the first channel of the M -channel audio signal contributes to the first fixed channel of the decorrelation input signal via a downmix signal in at least two coding formats. In other words, the first channel of the M -channel audio signal may contribute to the same channel of the decorrelation input signal via the downmix signal in each of these coding formats. It should be understood that in the present embodiment, in a specific coding format, the first channel of the M -channel audio signal may contribute, for example, to a plurality of decorrelation input signal channels via a downmix signal.

In the present exemplary embodiment, when switching said coding format between two coding formats, at least a portion of the first fixed channel of the decorrelation input signal remains. This may provide a smoother and/or less abrupt transition between coding formats for the listener's perception during playback of the reconstructed M -channel audio signal. In particular, the inventors have found that since a decorrelated signal can, for example, be created based on a part of the downmix signal corresponding to several time intervals during which switching between coding formats can occur in the downmix signal, as a result of switching between coding formats in the decorrelated audible distortion can potentially occur in the signal. Even if switching between coding formats interpolates wet and dry upmix coefficients, the reconstructed M -channel audio signal may still contain the distortion introduced in the decorrelated signal. Providing a decorrelation input signal according to the present embodiment can suppress such distortions in the decorrelated signal that are caused by switching between coding formats, which can improve the reproduction quality of the reconstructed M -channel audio signal.

In an exemplary embodiment, the pre-decrelation coefficients may be determined such that the second channel of the M -channel audio signal further contributes to the second fixed channel of the decorrelation input signal via a downmix signal in at least two coding formats. In other words, the second channel of the M -channel audio signal contributes to the same channel of the decorrelation input signal via a downmix signal in each of these coding formats. In the present exemplary embodiment, switching said coding format between two coding formats leaves at least a portion of the second fixed decorrelation input signal. Thus, the transition between encoding formats affects only one input of the decorrelator. This may provide a smoother and/or less abrupt transition between coding formats for the listener's perception during playback of the reconstructed M -channel audio signal.

The first and second channels of the M -channel audio signal may, for example, be different from each other. The first and second fixed channels of the decorrelation input signal may, for example, be different from each other.

In an exemplary embodiment, the received signaling may indicate a selected one of at least three coding formats, and the pre-decrelation coefficients may be determined such that the first channel of the M -channel audio signal contributes to the first fixed channel of the decorrelation input signal via the downmix signal in at least three encoding formats. In other words, the first channel of the M -channel audio signal contributes to the same channel of the decorrelation input signal through the downmix signal in these three coding formats. In the present exemplary embodiment, when switching said coding format between any of the three coding formats, at least a portion of the first fixed channel of the decorrelation input signal remains, which provides a smoother and/or less abrupt transition between coding formats for the listener to perceive during playback of the reconstructed M -channel audio signal.

In an exemplary embodiment, the pre-decrelation coefficients may be determined such that a channel pair of the M -channel audio signal contributes to a third fixed channel of the decorrelation input signal via a downmix signal in at least two coding formats. In other words, a pair of channels of an M -channel audio signal contributes to the same channel of the decorrelation input signal via a downmix signal in each of these coding formats. In the present embodiment, switching said coding format between the two coding formats leaves at least a portion of the third fixed channel of the decorrelation input signal, providing a smoother and/or less abrupt transition between coding formats for the listener to experience during playback of the reconstructed M -channel audio signal.

The channel pair may, for example, be different from the first and second channels of the M -channel audio signal. The third fixed channel of the decorrelation input may, for example, be different from the first and second fixed channels of the decorrelation input.

In an exemplary embodiment, the audio decoding method may further include performing a gradual transition from pre-decoration coefficient values associated with the first coding format to pre-decoration coefficient values associated with the second coding format, in response to determining whether said coding format has switched from the first coding format to the second format. coding. The use of a gradual transition between the values of the pre-decoration coefficients when switching between coding formats provides a smoother and/or less abrupt transition between coding formats for the listener's perception during playback of the reconstructed M -channel audio signal. In particular, the inventors have found that since a decorrelated signal can, for example, be created based on a part of the downmix signal corresponding to several time intervals during which switching between coding formats can occur in the downmix signal, as a result of switching between coding formats in the decorrelated audible distortion can potentially occur in the signal. Even if switching between coding formats interpolates wet and dry upmix coefficients, the reconstructed M -channel audio signal may still contain the distortion introduced in the decorrelated signal. Providing a decorrelation input signal according to the present embodiment can suppress such distortions in the decorrelated signal that are caused by switching between coding formats, which can improve the reproduction quality of the reconstructed M -channel audio signal.

The gradual transition may be performed, for example, by linear or continuous interpolation. The gradual transition can be performed, for example, by interpolation with a limited rate of change.

In an exemplary embodiment, the audio decoding method may further include interpolating from wet and dry upmix coefficient values, including coefficients with zero values, associated with the first coding format, to wet and dry upmix coefficient values, also including coefficients with zero values associated with the second coding format, in response to determining whether the specified coding format has switched from the first coding format to the second coding format. It should be understood that the downmix channels correspond to different combinations of channels from the originally encoded M -channel audio signal, so that an upmix factor with a value of zero in the first coding format will not necessarily have a value of zero in the second coding format, and vice versa. Preferably, the upmix coefficients are subject to interpolation rather than a compact representation of the coefficients, such as the representation described below.

Linear or continuous interpolation between the values of the upmix coefficients can be used, for example, to provide a smoother transition between coding formats for the listener's perception during playback of the reconstructed M -channel audio signal.

Large gradient interpolation, in which at a certain point in time associated with switching between coding formats, the old values of the upmix coefficients are replaced by new values of the upmix coefficients, can, for example, increase the fidelity of the reconstructed M -channel audio signal, for example, in cases of rapid change audio data of the M -channel audio signal and switching the encoding format on the encoder side in response to these changes to increase the fidelity of the reconstructed M -channel audio signal.

In an exemplary embodiment, the audio coding method may further include deriving signaling indicative of one of a plurality of interpolation schemes that may be used to interpolate wet and dry upmix parameters within a single coding format (i.e., by assigning the upmix coefficients new values during the period of time when there is no change in the encoding format), and the use of the specified interpolation scheme. Signaling indicative of one of a plurality of interpolation schemes may be obtained, for example, together with the downmix signal and/or the upmix parameters. Preferably, the interpolation scheme indicated by the signaling can additionally be used to transition between coding formats.

On the encoder side where the M -channel audio signal is available, it is possible, for example, to select interpolation schemes particularly suitable for the actual audio data of the M -channel audio signal. For example, linear or continuous interpolation can be used in cases where it is important for the overall impression of the reconstructed M -channel audio signal to ensure smooth switching, and interpolation with a large gradient, i.e., interpolation in which at a certain point in time associated with the transition between formats coding, old values of the upmix coefficients are replaced by new values of the upmix coefficients, can be used in cases where fast switching is important for the overall impression of the reconstructed M -channel audio signal.

In an exemplary embodiment, the at least two coding formats may include a first coding format and a second coding format. There is a gain factor that controls, in each coding format, the contribution from a channel of the M -channel audio signal to one of the line patterns corresponding to the channels of the downmix signal. In the present embodiment, the gain in the first coding format may be the same as the gain in the second coding format that controls the contribution from the same channel of the M -channel audio signal.

Using the same gain factors in the first and second coding formats can, for example, increase the similarity between the combined audio data of the downmix signal channels in the first coding format and the combined audio data of the downmix signal channels in the second coding format. Since the channels of the downmix signal are used to reconstruct the M -channel downmix signal, this can contribute to smoother transitions between the two coding formats for the listener's perception.

Using the same gain factors in the first and second coding formats can, for example, increase the similarity between the audio data of the first and second channels, respectively, of the downmix signal in the first coding format, and the audio data of the first and second channels, respectively, of the downmix signal in the second coding format. This may facilitate smoother transitions between the two encoding formats for the listener to perceive.

In the present embodiment, different gain factors can be used for different channels of the M -channel audio signal, for example. In the first example, the values of all gain factors in the first and second coding formats may be 1. In the first example, the first and second channels of the downmix signal may correspond to the unweighted sums of the first and second groups, respectively, in both the first and second coding formats. In the second example, the values of at least some of the gain factors may be other than 1. In the second example, the first and second channels of the downmix signal may correspond to the weighted sums of the first and second groups, respectively.

In an exemplary embodiment, the M -channel audio signal may comprise three channels representing different horizontal directions in the playback environment for the M -channel audio signal, and two channels representing directions that are vertically separated from the directions of the above three channels in the playback environment. In other words, an M -channel audio signal may comprise three channels intended to be reproduced by sound sources located substantially at the same height as the listener (or listener's ears) and/or propagating sound substantially horizontally, and two channels intended to be reproduced by sound sources located at a different height and/or propagating sound (substantially) non-horizontally. The two channels may, for example, represent directions from above.

In the exemplary embodiment, in the first coding format, the second channel group may comprise two channels representing directions vertically separated from the directions of the above three channels in the reproduction environment. By having these two channels in the second group, and using the same downmix signal channel to represent the two channels, it is possible, for example, to increase the fidelity of the reconstructed M -channel audio signal in cases where the vertical dimension in the playback environment is important to the overall M experience. - channel audio signal.

In the exemplary embodiment, in the first coding format, the first group of one or more channels may comprise three channels representing different horizontal directions in an M -channel audio playback environment, and the second group of one or more channels may comprise two channels representing directions separated vertically. from the directions of the above three channels in the reproduction environment. In the present embodiment, the first coding format allows three channels to be represented by the first channel of the downmix signal and two channels to be represented by the second channel of the downmix signal, whereby, for example, the reproduction fidelity of the reconstructed M -channel audio signal can be increased in cases where the vertical dimension is in playback environment is important to the overall M -channel audio experience.

In the exemplary embodiment, in the second coding format, each of the first and second groups may comprise one of two channels representing directions vertically separated from the directions of the above three channels in an M -channel audio reproduction environment. By having these two channels in different groups and using different channels of the downmix signal to represent the two channels, it is possible, for example, to increase the fidelity of the reconstructed M -channel audio signal in cases where the vertical dimension in the playback environment is not so important to the overall M-channel experience. channel audio.

In an exemplary embodiment, in a coding format referred to herein as a specific coding format, the first group of one or more channels may be composed of N channels, where N ≥ 3. In the present embodiment, if the specified coding format is a specific coding format, the pre-decorrelation coefficients may be determined such that, based on the first channel of the downmix signal, N − 1 channels of the decorrelated signal are generated; and the dry and wet upmix coefficients may be determined such that the first group of one or more channels is reconstructed as a linear mapping of the first channel of the downmix signal and N − 1 channels of the decorrelated signal, with a subset applied to the first channel of the downmix signal dry upmix coefficients, and a subset of wet upmix coefficients is applied to the N − 1 channels of the decorrelated signal.

The pre-decrelation coefficients may be determined, for example, such that the N − 1 channels of the decorrelation input signal coincide with the first channel of the downmix signal. The N − 1 channels of the decorrelated signal can be created, for example, by processing these N − 1 channels of the decorrelation input signal.

By reconstructing the first group of one or more channels as a linear mapping of the first channel of the downmix signal and N − 1 channels of the decorrelated signal, it is understood that the reconstructed version of the first group of one or more channels is obtained by linearly transforming the first channel of the downmix signal and N − 1 channels of the decorrelated signal signal. This linear transformation takes N channels as an input and provides N channels as an output, with a subset of dry upmix coefficients and a subset of wet upmix coefficients together constituting the coefficients quantifying this linear transformation.

In an exemplary embodiment, the obtained upmix parameters may include first type upmix parameters, referred to herein as "wet" upmix parameters, and second type upmix parameters, referred to herein as "dry" upmix parameters. In the present embodiment, determining sets of wet and dry upmix coefficients in a particular coding format may include determining a subset of dry upmix coefficients based on the dry upmix parameters; populating an intermediate matrix having more elements than the number of obtained wet upmix coefficients based on the obtained wet upmix coefficients and knowing that the intermediate matrix belongs to a class of predetermined matrices; and obtaining a subset of wet upmix coefficients by multiplying the intermediate matrix by a predetermined matrix, wherein the subset of wet upmix coefficients corresponds to the matrix resulting from the multiplication and includes more coefficients than the number of elements in the intermediate matrix.

In the present embodiment, the number of wet upmix coefficients in the subset of wet upmix coefficients is greater than the number of obtained wet upmix parameters. Using information about a given matrix and a class of given matrices to obtain a subset of wet upmix coefficients from the obtained wet upmix parameters can reduce the amount of information needed for parametric reconstruction of the first group from one or more channels, which can reduce the amount of metadata transmitted along with the downmix signal from the encoder side. Reducing the amount of data required for the parametric reconstruction makes it possible to reduce the required bandwidth for transmitting a parametric representation of an M -channel audio signal and/or the required storage space for such a representation.

A class of given matrices may be associated with known characteristics of at least some of the matrix elements that are valid for all matrices in the class, such as certain relationships between some of the matrix elements or zero values of some of the matrix elements. Knowing these characteristics allows the intermediate matrix to be populated based on fewer wet upmix parameters compared to the total number of matrix elements in the intermediate matrix. On the encoder side, information is available at least on the characteristics of the elements that are needed to calculate all elements of the matrix based on a smaller number of wet upmix coefficients, and on the relationships between these elements.

The definition and use of a given matrix and class of given matrices are described in more detail in U.S. Provisional Application No. 61/974,544, line 15 on page 16 to line 2 on page 20; Inventor named first: Lars Villemoes; filing date: April 3, 2014. In particular, see expression (9) in the aforementioned document as examples of a given matrix.

In an exemplary embodiment, the resulting upmix parameters may include N(N − 1)/2 wet upmix parameters. In the present embodiment, populating the intermediate matrix may include deriving values for (N − 1) ² matrix elements based on the N(N − 1)/2 obtained wet upmix parameters and knowing whether the intermediate matrix belongs to the class of given matrices. This may include entering wet upmix parameter values directly as matrix elements, or processing the wet upmix parameters appropriately to obtain values for matrix elements. In the present exemplary embodiment, the given matrix may include N(N − 1) elements, and the subset of wet upmix parameters may include N(N − 1) coefficients. For example, the resulting upmix parameters may include no more than N(N − 1)/2 independently assigned wet upmix parameters, and/or the number of wet upmix parameters may be no more than half the number of wet upmix parameters in a subset of wet upmix parameters.

In an exemplary embodiment, the obtained upmix parameters may include (N − 1) dry upmix parameters. In the present embodiment, the subset of dry upmix parameters may include N coefficients, and the subset of dry upmix parameters may be determined based on the obtained (N − 1) dry upmix parameters and based on a predetermined relationship between the coefficients in the subset dry” upmix parameters. For example, the resulting upmix parameters may include no more than (N − 1) independently assigned "dry" upmix parameters.

In an exemplary implementation, a class of given matrices may be one variant of upper or lower triangular matrices, where the known characteristics of all matrices in the class include zero elements of the given matrix; symmetric matrices, and the known characteristics of all matrices in the class include equal elements of a given matrix (on both sides of the main diagonal); and products of an orthogonal matrix and a diagonal matrix, wherein the known characteristics of all matrices in the class include known relationships between the elements of a given matrix. In other words, the class of given matrices can be the class of lower triangular matrices, the class of upper triangular matrices, the class of symmetric matrices, or the class of products of an orthogonal matrix and a diagonal matrix. A common property for each of the above classes is that their dimension is less than the total number of matrix elements.

In an exemplary embodiment, a given matrix and/or a class of given matrices can be associated with a given coding format, whereby, for example, the definition of the set of wet upmix parameters within the decoding method can be appropriately adapted.

In accordance with the exemplary embodiments, a method for decoding audio is provided, including receiving a signaling indicative of one of at least two given channel patterns; performing any of the audio decoding methods according to the first aspect in response to detecting received signaling indicative of the first predetermined channel pattern. A method for decoding audio in response to detecting received signaling indicative of a second predetermined channel pattern may include obtaining a two-channel downmix signal and associated upmix parameters; performing parametric reconstruction of the first three-channel audio signal based on the first channel of the downmix signal and at least some of the upmix parameters; and performing parametric reconstruction of the second three-channel audio signal based on the second channel of the downmix signal and at least some of the upmix parameters.

The first given channel configuration may correspond to an M -channel audio signal represented by the received two-channel downmix signal and associated upmix parameters. The second predetermined channel pattern may correspond to the first and second three-channel audio signals represented by the first and second channels of the received downmix signal, respectively, and the associated upmix parameters.

The ability to obtain signaling indicative of one of at least two given channel patterns and perform parametric reconstruction based on said channel pattern can provide a common format for a computer-readable medium containing a parametric representation of either an M -channel audio signal or two three-channel audio signals, from the side encoder on the decoder side.

According to exemplary embodiments, an audio decoding system is provided comprising a decoding section configured to reconstruct an M -channel audio signal based on a two-channel downmix signal and associated upmix parameters, with M ≥ 4. The audio decoding system comprises a control section configured to obtain signaling indicating a selected format from at least two M -channel audio coding formats. The coding formats are mapped to respective different channelizations of the M -channel audio signal into respective first and second groups of one or more channels. In this coding format, the first channel of the downmix signal corresponds to a linear combination of the first group of one or more channels of the M -channel audio signal, and the second channel of the downmix signal corresponds to the linear combination of the second group of one or more channels of the M -channel audio signal. The decoding section comprises a pre-decorrelation section configured to determine a set of pre-decorrelation coefficients based on a specified coding format and calculate an input decorrelation signal as a linear display of the downmix signal, wherein the set of pre-decorrelation coefficients is applied to the downmix signal; and a decorrelation section configured to generate a decorrelated signal based on the input decorrelation signal. The decoding section comprises a mixing section configured to determine sets of wet and dry upmix coefficients based on the obtained upmix parameters and a specified coding format; calculating the dry upmix signal as a linear display of the downmix signal, wherein a set of dry upmix coefficients is applied to the downmix signal; calculating the wet upmix signal as a linear display of the decorrelated signal, wherein a set of wet upmix coefficients is applied to the decorrelated signal; and combining dry and wet upmix signals to obtain a multi-dimensional reconstructed signal corresponding to the M -channel audio signal to be reconstructed.

In an exemplary embodiment, the audio decoding system may further comprise an additional decoding section configured to reconstruct the additional M -channel audio signal based on the additional two-channel downmix signal and the associated additional upmix parameters. The control system may be configured to receive signaling indicative of a selected format from at least two additional M -channel audio coding formats. The coding formats of the additional M -channel audio signal may be mapped to respective different channelizations of the additional M -channel audio signal into respective first and second groups of one or more channels. In this coding format of the additional M -channel audio signal, the first channel of the additional downmix signal may correspond to a linear combination of the first group of one or more channels of the additional M -channel audio signal, and the second channel of the additional downmix signal may correspond to a linear combination of the second group of one or more channels of the additional M - channel audio signal. The additional decoding section may comprise an additional pre-decorrelation section configured to determine an additional set of pre-decoration coefficients based on the specified coding format of the additional M -channel audio signal and calculate the additional decorrelation input as a linear mapping of the additional downmix signal, wherein the additional set is applied to the additional downmix signal. pre-decoration coefficients; and an additional decorrelation section configured to generate an additional decorrelated signal based on the additional decorrelation input signal. The additional decoding section may further comprise an additional mixing section configured to determine additional sets of wet and dry upmix coefficients based on the additional upmix parameters obtained and the specified additional M -channel audio coding format; calculating the additional dry upmix signal as a linear mapping of the additional downmix signal, wherein an additional set of dry upmix coefficients is applied to the additional downmix signal; calculating the additional wet upmix signal as a linear mapping of the additional decorrelated signal, wherein an additional set of wet upmix coefficients is applied to the additional decorrelated signal; and combining the additional dry and wet upmix signals to obtain an additional multi-dimensional reconstructed signal corresponding to the additional M -channel audio signal to be reconstructed.

In the present embodiment, the additional decoding section, the additional pre-decorrelation section, the additional decorrelation section, and the additional mixing section may be configured to operate independently of the decoding section, the pre-decorrelation section, the decorrelation section, and the mixing section, for example.

In the present embodiment, the additional decoding section, the additional pre-decorrelation section, the additional decorrelation section, and the additional mixing section may be, for example, functionally equivalent to the decoding section, the pre-decorrelation section, the decorrelation section, and the mixing section, respectively (or similar). In an alternative embodiment, at least one of the additional decoding section, the additional pre-decorrelation section, the additional decorrelation section, and the additional mixing section can be performed, for example, with the possibility of performing at least one type of interpolation different from that performed by the decoding section, the pre-decorrelation section, the section decorrelation and mixing section.

For example, the received signaling may indicate different coding formats for the M -channel audio signal and the additional M -channel audio signal. In an alternative embodiment, the coding formats of the M -channel audio signals may, for example, always match, and the received signaling may indicate a selected format from at least two common coding formats for the two M -channel audio signals.

The interpolation schemes used for fading between pre-decoration coefficients in response to switching between M -channel audio coding formats may be the same as the interpolation schemes used for fading between additional pre-decorrelation coefficients in response to switching between additional M -channel audio coding formats, or may differ from these schemes.

Similarly, the interpolation schemes used to interpolate the wet and dry upmix coefficient values in response to switching between M -channel audio coding formats may be the same as the interpolation schemes used to interpolate the additional wet and dry upmix coefficient values. mixing in response to switching between additional M -channel audio coding formats, or may differ from these schemes.

In an exemplary embodiment, the audio decoding system may further comprise a demultiplexer configured to extract from the bitstream of the downmix signal, the upmix parameters associated with the downmix signal, and the discretely encoded audio channel. The decoding system may further comprise a single-channel decoding section configured to decode the discretely encoded audio channel. The discretely encoded audio channel may be, for example, bitstream encoded with a perceptual audio codec such as Dolby Digital, MPEG AAC, or derivatives thereof, and the single-channel decoding section may comprise, for example, a core decoder for decoding the discretely encoded audio channel. The single-channel decoding section may, for example, be configured to decode the discrete-coded audio channel independently of the decoding section.

In accordance with the exemplary embodiments, a computer program product is provided, comprising a computer-readable medium with instructions for carrying out any of the methods of the first aspect.

II. Review. encoder side

According to a second aspect, the exemplary embodiments provide audio coding systems as well as audio coding methods and associated computer software products. The proposed coding systems, methods, and computer program products according to the second aspect may have essentially the same general features and benefits. Moreover, the advantages presented above for the features of the coding systems, methods, and computer program products according to the first aspect may be substantially valid for the corresponding features of the coding systems, methods, and computer program products according to the second aspect.

According to exemplary embodiments, an audio coding method is provided, including obtaining an M -channel audio signal for which M ≥ 4. The audio coding method includes repeatedly selecting one format from at least two coding formats based on any suitable selection criteria, such as signal characteristics, loading system, user preferences, network conditions. The selection may be repeated once for each time slot of the audio signal, or once for every ^nth time slot, which may result in a different format being selected than originally selected; in an alternative embodiment, the selection may be event driven. The coding formats are mapped to respective different channelizations of the M -channel audio signal into respective first and second groups of one or more channels. In each of the coding formats, the two-channel downmix signal includes a first channel formed as a linear combination of the first group of one or more channels of the M -channel audio signal, and a second channel formed as a linear combination of the second group of one or more channels of the M -channel audio signal. For the selected coding format, the downmix channel is calculated based on the M -channel audio signal. After the downmix signal of the currently selected coding format is calculated, a signaling indicating the currently selected coding format and additional information is output, thereby allowing parametric reconstruction of the M -channel audio signal. If the selection results in switching the first selected coding format to a second (different) selected coding format, a transition can be triggered, resulting in a smooth transition output of the downmix signal according to the first selected coding format and the downmix signal according to the second selected format. coding. In this case, the smooth transition may be a linear or non-linear interpolation of the two signals. For example, the expression

y(t) = tx ₁ (t) + (1 − t)x ₂ (t), t ∈ [0,1]

provides a smooth transition y from function x ₂ to function x ₁ with a linear dependence on time, and x ₁ , x ₂ can be vector-valued functions of time representing the downmix signals in accordance with the respective coding formats. To simplify the display, the time interval over which the fade occurs has been scaled to [0, 1], where t = 0 represents the start of the fade and t = 1 represents the point in time when the fade is completed.

The location of times t = 0 and t = 1 in physical units can be important to the perceived output quality of the reconstructed audio. As a guideline for the location of the fade, it can start as early as possible after the need to change the format is determined, and/or finish the fade as soon as possible so that it is imperceptible to perception. Thus, for implementations in which the choice of coding format is repeated every time interval, according to some embodiments, the soft transition is started (t = 0) at the beginning of the time interval, and the end point of the soft transition (t = 1) is located as far as possible. closer, yet far enough away that the average listener cannot discern distortion or degradation due to the transition between two reconstructions of a common M -channel audio signal (with normal content) based on two different coding formats. In one embodiment, the downmix signal output from the audio coding method is segmented into time slots, and the fade may span one slot. In another embodiment, the downmix signal output from the audio coding method is segmented into overlapping time slots, and the duration of the fade corresponds to a step from one time slot to another.

In exemplary embodiments, signaling indicative of the currently selected coding format may be encoded based on consecutive time slots. In an alternative embodiment, the signaling may be time-differentiated in the sense that such signaling may not be activated for one or more consecutive time slots unless the selected coding format changes. At the decoder side, such a sequence of slots can be interpreted such that the most recently signaled coding format remains selected.

Depending on the audio data of the M -channel audio signal, various divisions of the channels of the M -channel audio signal into first and second groups, represented by the corresponding downmix signal channels. Therefore, the reliability of the reconstructed M -channel audio signal can be improved by selecting an appropriate coding format, namely the best fit among the given coding formats.

In the exemplary embodiment, the additional information includes "dry" and "wet" upmix coefficients in the same sense as these terms have been used in the present application above. If we are not talking about the tasks of a special implementation, it is usually sufficient to calculate additional information (in particular, "dry" and "wet" upmix coefficients) for the currently selected encoding format. In particular, a set of dry upmix coefficients (which may be represented as an M ×2 matrix) may define a linear mapping of the corresponding downmix signal approximating an M -channel audio signal. A set of wet upmix coefficients (which can be represented as an M × P matrix, where the number of decorrelators P can be P = M − 2) defines a linear mapping of the decorrelated signal, so that the covariance of the signal obtained by the specified linear mapping of the decorrelated signal , complements the covariance of the M -channel audio signal approximated by a linear mapping of the downmix signal of the selected coding format. Displaying the decorrelated signal, which is defined by a set of wet upmix coefficients, will complement the covariance of the M -channel (approximated) audio signal in that the covariance of the sum of the M -channel audio signal and the display of the decorrelated signal is usually closer to the covariance of the received M -channel audio signal. The effect of introducing additional covariance may be to increase the reliability of the reconstructed signal at the decoder side.

The linear display of the downmix signal provides an approximation of an M -channel audio signal. In the reconstruction of the M -channel audio signal at the decoder side, the decorrelated signal is used to increase the dimension of the audio data of the downmix signal, and the signal obtained by linear mapping of the decorrelated signal is combined with the signal obtained by linear mapping of the downmix signal to increase the approximation reliability of the M -channel audio signal. . Since the decorrelated signal is determined based on at least one channel of the downmix signal and the decorrelated signal does not contain any audio data from the M -channel audio signal that is not yet available in the downmix signal, the difference between the covariance of the obtained M -channel audio signal and the covariance of M - of a channel audio signal approximated by a linear display of the downmix signal can indicate not only the validity of an M -channel audio signal approximated by a linear display of the downmix signal, but also the validity of an M -channel audio signal reconstructed by both the downmix signal and decorrelated signal. In particular, the reduced difference between the covariance of the received M -channel audio signal and the covariance of the M -channel audio signal approximated by linear mapping of the downmix signal may indicate increased fidelity of the reconstructed M -channel audio signal. The display of the decorrelated signal, which is given by a set of wet upmix coefficients, complements the covariance of the M -channel audio signal (derived from the downmix signal) in that the covariance of the sum of the M -channel audio signal and the display of the decorrelated signal is closer to the covariance of the received M -channel audio signal . Therefore, the choice of one of the coding formats based on the respective calculated differences can increase the reliability of the reconstructed M -channel audio signal.

It should be understood that the coding format may be selected, for example, directly based on the calculated differences, or on the basis of coefficients and/or values determined based on the calculated differences.

In addition, it should be understood that the coding format can be selected, for example, based on the corresponding calculated "dry" upmix coefficients in addition to the corresponding calculated differences.

A set of "dry" upmix coefficients can be determined, for example, by a minimum standard error approximation, assuming that only the downmix signal is available for reconstruction, i.e., assuming that no decorrelated signal is used for reconstruction.

The calculated differences may be, for example, the differences between the obtained M -channel audio covariance matrix and the M -channel audio covariance matrices approximated by respective downmix signal line mappings of different coding formats. Selecting one of the encoding formats may include, for example, calculating matrix norms for the respective differences between the covariance matrices, and selecting one of the encoding formats based on the calculated matrix norms, for example, selecting an encoding format associated with the minimum of the calculated matrix norms.

A decorrelated signal may, for example, include at least one channel and at most M − 2 channels.

By a set of dry upmix coefficients defining a linear mapping of the downmix signal approximating the M -channel downmix signal, it is meant that the approximation of the M -channel downmix signal is carried out by linear transformation of the downmix signal. This linear conversion takes two channels of a downmix signal as an input and provides M channels as an output, and the dry upmix coefficients are the coefficients quantifying this linear conversion.

Similarly, the "wet" upmix parameters define the quantitative characteristics of the linear transformation, in which the channels (s) of the decorrelated signal are taken as an input and provide M channels as an output.

In an exemplary embodiment, the wet upmix parameters may be determined such that the covariance of the signal obtained by linear mapping (which is given by the wet upmix parameters) of the decorrelated signal approximates the difference between the covariance of the received M -channel audio signal and the covariance of the M -channel audio signal, approximated by a linear mapping of the downmix signal of the selected coding format. In other words, the covariance of the sum of the first linear mapping (defined by the dry upmix parameters) of the downmix signal and the second linear mapping (defined by the wet upmixing parameters defined according to this embodiment) of the decorrelated signal will be close to the covariance of the M -channel an audio signal expressing input to the audio coding method described above in this application. Determining wet upmix coefficients in accordance with the present embodiment can increase the fidelity of the reconstructed M -channel audio signal.

In an alternative embodiment, the wet upmix parameters may be determined such that the covariance of the signal obtained by linearly mapping the decorrelated signal approximates a portion of the difference between the covariance of the received M -channel audio signal and the covariance of the M -channel audio signal approximated by linearly mapping the downmix signal of the selected encoding format. If, for example, a limited number of decorrelators are available at the decoder side, it may not be possible to completely reconstruct the covariance of the received M -channel audio signal. In such an example, wet upmix parameters suitable for partial reconstruction of the M -channel audio covariance using fewer decorrelators can be determined at the encoder side.

In an exemplary embodiment, the audio coding method may further include, for each of the at least two coding formats, defining a set of wet upmix coefficients that, together with dry upmix coefficients (of that coding format), allow parametric reconstruction of an M -channel audio signal from a downmix signal (of this coding format) and from a decorrelated signal determined based on the downmix signal (of this format), wherein the set of wet upmix coefficients defines a linear mapping of the decorrelated signal such that the covariance of the signal obtained by linear mapping of the decorrelated signal, approximated the difference between the covariance of the received M -channel audio signal and the covariance of the M -channel audio signal approximated by a linear display of the downmix signal (of this format). In the present exemplary embodiment, the selected coding format may be selected based on the values of the respective specific sets of wet upmix coefficients.

An indication of the validity of the reconstructed M -channel audio signal can be obtained, for example, based on certain "wet" upmix coefficients. The choice of coding format may be based, for example, on the weighted or unweighted sums of the determined wet upmix coefficients, on the weighted or unweighted sums of the moduli of the determined wet upmix coefficients, and/or on the weighted or unweighted sums of the squares of the determined wet upmix coefficients. mixing, in addition, for example, on the respective sums of the respective calculated "dry" upmix coefficients.

The wet upmix parameters may, for example, be computed for a plurality of M -channel signal bands, and the choice of coding format may be based, for example, on the values of respective specific sets of wet upmix coefficients in the respective frequency bands.

In an exemplary embodiment, the transition between the first and second coding formats includes outputting discrete values of dry and wet upmix coefficients of the first coding format at one time interval and the second coding format at a subsequent time interval. The functions of the decoder that ultimately provide reconstruction of the M -channel signal may include the interpolation of the upmix coefficients between the output samples. With such functions on the decoder side, a smooth transition from the first coding format to the second can be effectively ensured. As with the smooth transition used for the downmix signal as described above, such a smooth transition may result in less likelihood of perceiving a transition between coding formats when reconstructing an M -channel audio signal.

It will be understood that the coefficients used to calculate the downmix signal based on the M -channel audio signal can be interpolated, i.e., from values associated with the time interval over which the downmix signal is calculated according to the first coding format, to values associated with a time interval at which the downmix signal is calculated according to the second coding format. At least if the downmix occurs in the time domain, the soft transition of the downmix as a result of coefficient interpolation will be equivalent to the smooth transition of the downmix as a result of coefficient interpolation performed directly on the respective downmix signals. It should be understood that the coefficient values used to calculate the downmix signal are typically independent of the signals, but may be specified for each of the available coding formats.

When it comes to the fading of the downmix signal and the upmix coefficients, the advantage is that synchronization between the two fadings is ensured. Preferably, it is possible to match the respective transition periods for the downmix signal and the upmix coefficients. In particular, the means responsible for the respective smooth transitions can be controlled by the overall flow of control data. Such control data may include soft transition start and end points, and optionally a soft transition curve such as linear, non-linear curve, etc. Regarding the upmix coefficients, the soft transition curve may be set by a predetermined interpolation rule that determines the mode of the decoding apparatus; however, the start and end points of smooth transitions can be controlled indirectly by the positions at which the discrete values of the upmix coefficients are set and/or output. The similarity of the time dependence of the two smooth transition processes provides a high degree of correspondence between the downmix signal and the parameters provided for its reconstruction, which can lead to reduced distortion on the decoder side.

In an exemplary embodiment, the choice of coding format is based on a comparison of the difference in covariance between the received M -channel signal and the reconstructed M -channel signal from the downmix signal. In particular, the reconstruction may coincide with a linear display of the downmix signal, defined only by "dry" upmix coefficients, i.e., without the contribution from the signal, determined by decorrelation (for example, to increase the dimension of the audio data of the downmix signal). In particular, the comparison should not take into account any contribution from the linear mapping defined by any set of wet upmix coefficients. In other words, the comparison is carried out under the condition that the decorrelated signal is not available. This principle can help select an encoding format that allows for more accurate reproduction. After performing this comparison and deciding on the choice of coding format, a set of "wet" upmix coefficients is optionally determined. An advantage associated with this process is that there is no double determination of wet upmix coefficients for a particular portion of the received M -channel audio signal.

According to a modification of the exemplary embodiment described in the previous paragraph, dry and wet upmix coefficients are calculated for all coding formats, and a quantitative measure of the wet upmix coefficients is used as the basis for selecting the coding format. Indeed, a score calculated from certain wet upmix coefficients may provide an (inverse) indication of the validity of the reconstructed M -channel audio signal. The choice of coding format may be based, for example, on the weighted or unweighted sums of the determined wet upmix coefficients, on the weighted or unweighted sums of the moduli of the determined wet upmix coefficients, and/or on the weighted or unweighted sums of the squares of the determined wet upmix coefficients. mixing. Each of these options can be used in conjunction with the respective sums of the respective calculated "dry" upmix coefficients. The wet upmix parameters may, for example, be computed for a plurality of M -channel signal bands, and the choice of coding format may be based, for example, on the values of respective specific sets of wet upmix coefficients in the respective frequency bands.

In an exemplary embodiment, the audio coding method may further include calculating the sum of squares of respective wet upmix coefficients and the sum of squares of respective dry upmix coefficients for each of the at least two coding formats. In the present embodiment, the selected coding format may be selected based on sums of squares. The inventors have found that the computed sums of squares can provide a particularly accurate indication of the loss of listener confidence that will occur when an M -channel audio signal is reconstructed based on the mixed wet and dry signal contributions.

For example, for each coding format, a ratio can be created based on the computed sums of squares for the corresponding coding format, and the selected format can be associated with the minimum or maximum of the generated ratios. The creation of the ratio may include, for example, dividing, on the one hand, the sum of the squares of the wet upmix coefficients by, on the other hand, the sum of the sum of the squares of the dry upmix coefficients and the sum of the squares of the wet upmix coefficients. In an alternative embodiment, the ratio can be created by dividing the sum of squares of the wet upmix coefficients by the sum of the squares of the dry upmix coefficients.

In an exemplary embodiment, the method proposes encoding an M -channel audio signal and at least one associated ( M ₂ -channel) audio signal. Audio signals can be linked in the sense that they can describe common audio data, for example, when recorded simultaneously or generated during the same audio data creation process. The audio signals need not be encoded with a common downmix signal, but may be encoded with separate processes. With such parameters, the choice of one of the coding formats additionally takes into account the data related to the specified at least one additional audio channel, and the coding format is thus chosen to be used for encoding both the M -channel audio signal and the associated ( M ₂ -channel) audio signal.

In the exemplary embodiment, the downmix signal output by the audio coding method may be segmented into time slots, the coding format selection may be performed once per time slot, and the selected coding format may be maintained for at least a given number of time slots until another format is selected. coding. The choice of coding format for a timeslot can be done in any of the following ways, such as by considering differences between covariances, considering wet upmix coefficient values for available coding formats, and the like. Maintaining the chosen coding format for a minimum number of timeslots can allow, for example, avoid repeated switching between encoding formats. In the present exemplary embodiment, it is possible, for example, to improve the reproduction quality of the reconstructed M -channel audio signal as perceived by the listener.

The minimum number of timeslots can be, for example, 10.

The received M -channel audio signal, for example, can be buffered for a minimum number of time slots, and the choice of coding format can be made, for example, based on a majority decision over a moving window containing a number of time slots, selected taking into account the specified minimum number of time slots, in order to support selected encoding format. The implementation of such a stabilizing function may include one of a variety of anti-aliasing filters, in particular finite impulse response anti-aliasing filters known in the field of digital signal processing. As an alternative to this approach, the coding format may be switched to a new coding format upon detecting that a new coding format has been selected in order for a specified minimum number of time slots. To implement this criterion, a moving time window with a minimum number of consecutive time slots can be applied to past coding format choices, such as for buffered time slots. If, after the sequence of slots of the first coding format, for each slot in the moving window, the second coding format remains selected, then the transition to the second coding format, which takes effect from the beginning of the moving window onwards, is confirmed. The implementation of the above stabilizing function may include a state machine.

In an exemplary embodiment, a compact representation of dry and wet upmix parameters is proposed, which includes, among other things, the creation of an intermediate matrix, which, due to belonging to a class of given matrices, is uniquely determined by fewer parameters compared to the number of elements in the matrix. Aspects of this compact representation have been described above in the present description and, in particular, with reference to US Provisional Application No. 61/974,544, the author, whose name is listed first: Lars Villemoes; filing date: April 3, 2014

In an exemplary embodiment, in the selected coding format, the first group of one or more channels of the M -channel audio signal may consist of N channels, where N ≥ 3. The first group of one or more channels may be characterized by the possibility of reconstruction from the first channel of the downmix signal and N − 1 channels of decorrelated signal by applying at least wet and dry upmix coefficients.

In the present embodiment, determining a set of dry upmix coefficients of the selected coding format may include determining a subset of dry upmix coefficients of the selected coding format to determine a linear mapping of the first channel of the downmix signal of the selected coding format to approximate a first group of one or more channels of the selected encoding format.

In the present embodiment, determining the set of wet upmix coefficients of the selected coding format may include determining an intermediate matrix based on the difference between the first group covariance from the received one or more channels of the selected coding format and the covariance of the first group from one or more channels of the selected coding format approximated by by linearly displaying the first channel of the downmix signal of the selected coding format. The product of the intermediate matrix and the given matrix may correspond to a subset of the wet upmix coefficients of the selected coding format, resulting in a linear mapping of N − 1 channels of the decorrelated signal as part of a parametric reconstruction of the first group from one or more channels of the selected coding format. The subset of wet upmix coefficients of the selected coding format may include more coefficients than the number of elements in the intermediate matrix.

In the present exemplary embodiment, the output upmix parameters may include a set of first type upmix parameters, referred to herein as "dry" upmix parameters, from which a subset of "dry" upmix parameters can be extracted, and a set of second type upmix parameters, referred to in the present application as the present application with "wet" upmix parameters that uniquely define an intermediate matrix, provided that it belongs to the class of given matrices. The intermediate matrix may have more elements compared to the number of elements in the wet upmix subset of the selected coding format.

In the present embodiment, a copy of the parametric reconstruction of the first group of one or more channels on the decoder side includes, as one contribution, a "dry" upmix signal generated by linear mapping of the first channel of the downmix signal, and, as an additional contribution, " a wet" upmix signal resulting from a linear mapping of N − 1 channels of a decorrelated signal. A subset of dry upmix coefficients defines a linear mapping of the first channel of the downmix signal, and a subset of wet upmix coefficients defines a linear mapping of the decorrelated signal. As a result of issuing wet upmix parameters, the number of which is less than the number of coefficients in the subset of wet upmix coefficients and from which the subset of wet upmix coefficients can be extracted based on the predetermined matrix and the class of predetermined matrices, it is possible to reduce the amount of information transmitted to the decoder side to enable reconstruction of the M -channel audio signal. Reducing the amount of data required for the parametric reconstruction makes it possible to reduce the required bandwidth for transmitting a parametric representation of an M -channel audio signal and/or the required storage space for such a representation.

The intermediate matrix can be defined, for example, such that the covariance of a signal obtained by linearly mapping the N − 1 channels of the decorrelated signal complements the covariance of the first group of one or more channels approximated by linearly mapping the first channel of the downmix signal.

The definition and use of a given matrix and a class of given matrices are described in more detail in the aforementioned provisional application for US patent No. 61/974,544 from line 15 on page 16 to line 2 on page 20. In particular, as examples of a given matrix, see expression (9) in the above document.

In an exemplary embodiment, the definition of the intermediate matrix may include defining the intermediate matrix such that the covariance of the signal obtained by linearly mapping N − 1 channels of the decorrelated signal determined by a subset of wet upmix coefficients approximates the difference between the covariance of the first group of the obtained one or more channels and the covariance of the first group of one or more channels approximated by a linear mapping of the first channel of the downmix signal, or essentially the same as this difference. In other words, the intermediate matrix may be defined such that a reconstruction copy of the first group of one or more channels obtained as the sum of the dry upmix signal generated by linearly mapping the first channel of the downmix signal and the wet upmix signal generated by linear mapping of the N − 1 channels of the decorrelated signal completely or at least approximately restores the covariance of the first group of the received one or more channels.

In an exemplary embodiment, the wet upmix parameters may include at most N(N − 1)/2 independently assigned wet upmix parameters. In the present embodiment, the intermediate matrix may have (N − 1) ² matrix elements and may be uniquely determined by wet upmix parameters, provided that the intermediate matrix belongs to the class of given matrices. In the present exemplary embodiment, the subset of wet upmix coefficients may include N(N − 1) coefficients.

In an exemplary embodiment, the subset of dry upmix coefficients may include N coefficients. In the present embodiment, the dry upmix parameters may include at most N − 1 dry upmix parameters, and a subset of the dry upmix coefficients may be extracted from the N − 1 dry upmix parameters by a predetermined rule.

In an exemplary embodiment, a certain subset of dry upmix coefficients may define a linear mapping of the first channel of the downmix signal corresponding to the minimum standard error approximation of the first group of one or more channels, i.e., the set of linear mappings of the first channel of the downmix signal, wherein a certain set of "dry" upmix coefficients may define a linear mapping that best approximates the first group of one or more channels with a minimum mean square error.

In an exemplary embodiment, an audio coding system is provided, comprising an encoding section configured to encode an M -channel audio signal as a two-channel audio signal and associated upmix parameters, with M ≥ 4. The encoding section comprises a downmix section configured to compute a two-channel downmix signal based on the M -channel audio signal for at least one of at least two coding formats mapped to different channelizations of the M -channel audio signal into respective first and second groups of one or more channels, in accordance with the coding format. The first channel of the downmix signal is formed as a linear combination of the first group of one or more channels of the M -channel audio signal, and the second channel of the downmix signal is formed as a linear combination of the second group of one or more channels of the M -channel audio signal.

The audio encoding system further comprises a control section configured to select one of the encoding formats based on any suitable criteria such as signal characteristics, system loading, user preferences, network conditions. The audio coding system further comprises a downmix interpolator configured to make smooth transitions of the downmix signal between the two coding formats upon a transition request from the control section. During such a transition, it is possible to compute downmix signals for both coding formats. In addition to outputting the downmix signal or, when applicable, fading it, the audio coding system is configured to output at least a signaling indicative of the currently selected coding format and additional information, thereby enabling parametric reconstruction M - channel audio signal based on the downmix signal. If the system contains a plurality of coding sections that operate in parallel, for example, to encode respective groups of audio channels, then the control section may be implemented to operate independently of each of them and may be responsible for selecting a common coding format to be used by each of the coding sections.

In an exemplary embodiment, a computer program product is provided, comprising a computer readable medium with instructions for performing any of the methods described in this section.

III. Implementation examples

On FIG. 6-8 show alternative ways of dividing 11.1-channel audio signal into channel groups for parametric encoding of 11.1-channel audio signal as 5.1-channel audio signal. 11.1-channel audio contains L (Left), LS (Surround Left), LB (Surround Left), TFL (Front Top Left), TBL (Surround Back Left), R (Right), RS (Right Side), RB ( right back), TFR (top front right), TBR (top back right), C (center), and LFE (low frequency effects channel). The five channels L, LS, LB, TFL, and TBL form a five-channel audio signal representing the left half of space in an 11.1-channel audio playback environment. The three channels L, LS and LB represent different horizontal directions in the playback environment, and the two channels TFL and TBL represent directions separated vertically from the three channels L, LS and LB. The two channels TFL and TBL can be used, for example, for playback in ceiling loudspeakers. Similarly, five channels R, RS, RB, TFR and TBR form an additional five-channel audio signal representing the right half in the playback environment, with three channels R, RS and RB representing different horizontal directions in the playback environment, and two channels TFR and TBR representing directions separated vertically from the three channels R, RS and RB.

To represent 11.1-channel audio as 5.1-channel audio, the set of L, LS, LB, TFL, TBL, R, RS, RB, TFR, TBR, C, and LFE channels can be divided into channel groups represented by the respective downmix channels and associated upmix settings. The five-channel audio signal L, LS, LB, TFL, TBL can be represented by a two-channel downmix signal L ₁ , L ₂ and associated upmix parameters, and the five-channel audio signal R, RS, RB, TFR, TBR can be represented by an additional two-channel downmix signal R ₁ , R ₂ and associated additional upmix parameters. The C and LFE channels can also be separated in 5.1-channel representation of 11.1-channel audio.

On FIG. 6 shows a first coding format F ₁ in which a five-channel audio signal L, LS, LB, TFL, TBL is divided into a first L, LS, LB channel group 601 and a second TFL, TBL channel group 602, and in which an additional five-channel audio signal R, RS, RB, TFR, TBR is divided into an optional first R, RS, RB channel group 603 and an optional second TFR, TBR channel group 604. In the first F ₁ coding format, the first channel group 601 is represented by the first channel L ₁ of the two-channel downmix signal, and the second channel group 602 is represented by the second channel L ₂ of the two-channel downmix signal. The first channel L ₁ of the downmix signal may correspond to the sum of the channels of the first group 601, i.e., L ₁ = L + LS + LB, and the second channel L ₂ of the downmix signal may correspond to the sum of the channels of the second group 602, i.e., L ₂ = TFL + TBL.

In exemplary embodiments, some or all of the channels may be pre-stacked so that the first channel L ₁ of the downmix signal may correspond to a linear combination of the first channel group 601, i.e., L ₁ = c ₁ L + c ₂ LS + c ₃ LB, and the second channel L ₂ of the downmix signal could correspond to a linear combination of the second channel group 602, i.e., L ₂ = c ₄ TFL + c ₅ TBL. The gains c ₂ , c ₃ , c ₄ , c ₅ , for example, may be the same, and the gain c ₁ , for example, may have a different value; for example, c ₁ could correspond to no scaling at all. For example, you can use the values c ₁ = 1 and c ₂ = c ₃ = c ₄ = c ₅ = 1/√2. If, for example, the gains c ₁ , ..., c ₅ that apply to the respective channels L, LS, LB, TFL, TBL in the first coding format F ₁ coincide with the gains that apply to these channels in other formats coding F ₂ and F ₃ , described below with respect to FIG. 7 and 8, these gains do not affect how the downmix signal changes when switching between different coding formats F ₁ , F ₂ , F ₃ , but the scaled channels c ₁ L, c ₂ LS, c ₃ LB, c ₄ TFL , c ₅ TBL, therefore, can be processed in the same way as if they were the original channels L, LS, LB, TFL, TBL. On the other hand, if different gain factors act to change the scale of the same channels in different coding formats, then switching between these coding formats can lead, for example, to transitions between L, LS, LB, TFL, TBL channel variants with different scaling in the signal downmix, which can lead to audible distortion on the decoder side. Such distortion can, for example, be suppressed by interpolating from the coefficients used to create the downmix signal before switching the coding format to the coefficients used to create the downmix signal after switching the coding format, and/or by applying interpolation of pre-decorrelation coefficients, as described below. in connection with expressions (3) and (4).

Similarly, the additional first channel group 603 is represented by the first channel of the additional downmix signal R ₁ and the additional second group of channels 604 is represented by the second channel of the additional downmix signal R ₂ .

The first coding format F ₁ provides downmix channels L ₂ and R ₂ specifically designed to represent overhead channels TFL, TBL, TFR and TBR. Therefore, the use of the first coding format F ₁ can enable the parametric reconstruction of the 11.1-channel audio signal with a relatively high degree of certainty in cases where, for example, in a playback environment, the vertical dimension is important for the overall perception of the 11.1-channel audio signal.

On FIG. 7 shows a second coding format F ₂ in which a five-channel audio signal L, LS, LB, TFL, TBL is divided into a first channel group 701 and a second channel group 702, which are represented by the respective channels of the downmix signal L ₁ , L ₂ , wherein the channels L ₁ and L ₂ correspond to the sums of the respective channel groups 701 and 702 or linear combinations of the respective channel groups 701 and 702 using the same gains c ₁ , ..., c ₅ to scale the respective channels L, LS, LB, TFL, TBL, which and in the first coding format F ₁ . Similarly, the additional five-channel audio signal R, RS, RB, TFR, TBR is divided into additional first channel group 703 and second channel group 704, which are represented by the respective channels R ₁ and R ₂ .

The second coding format F ₂ does not provide downmix channels specifically designed to represent the overhead channels TFL, TBL, TFR and TBR, but may allow relatively high fidelity parametric reconstruction of 11.1-channel audio in cases where, for example, in a playback environment for the overall impression of 11.1-channel audio, the vertical dimension is not so important.

On FIG. 8 shows a third coding format F ₃ in which a five-channel audio signal L, LS, LB, TFL, TBL is divided into a first group 801 and a second group 802 of one or more channels, which are represented by the corresponding channels L ₁ and L ₂ of the downmix signal, wherein channels L ₁ and L ₂ correspond to sums of respective groups 801 and 802 from one or more channels, or linear combinations of respective groups 801 and 802 from one or more channels, using the same coefficients c ₁ , ..., c ₅ to rescale the respective channels L, LS, LB, TFL, TBL, as in the first coding format F ₁ . Similarly, the additional five-channel signal R, RS, RB, TFR, TBR is divided into additional first channel group 803 and second channel group 804, which are represented by the respective channels R ₁ and R ₂ . In the third coding format F _{3 ,} only the channel L is represented by the first channel L ₁ of the downmix signal, and the four channels LS, LB, TFL and TBL are represented by the second channel L ₂ of the downmix signal.

As described with respect to FIG. 1 to 5, the encoder side calculates the two-channel downmix signal L ₁ , L ₂ as a linear mapping of the five-channel audio signal X = [L LS LB TFL TBL] ^T according to the expression

where d _n,m , with n = 1, 2, m = 1...,5, are the downmix coefficients represented by the downmix matrix D. As described with respect to FIG. 9-13, on the decoder side, the parametric reconstruction of the five-channel audio signal [L LS LB TFL TBL] ^T is performed according to the expression

where c _n,m , for n = 1, ..., 5, m = 1, 2, are the dry upmix coefficients represented by the dry upmix matrix β _L , p _n,k , for n = 1, ..., 5, k = 1, 2, 3 are the “wet” upmix coefficients represented by the “wet” upmix matrix γ _L , and z _k , at k = 1, 2, 3 are the channels of the three-channel decorrelated signal Z, created on the basis of the downmix signal L ₁ , L ₂ .

On FIG. 1 is a generalized block diagram of an encoding section 100 for encoding an M -channel audio signal as a two-channel downmix signal and associated upmix parameters in accordance with an embodiment.

In the present application, an example of an M -channel audio signal is the five-channel audio signal L, LS, LB, TFL, and TBL described with respect to FIG. 6–8. In addition, embodiments may be provided in accordance with which the encoding section 100 is configured to calculate a two-channel downmix signal based on an M -channel audio signal, where M = 4 or M ≥ 6.

The encoding section 100 contains a downmix section 110 and an analysis section 120 . For each of the coding formats F ₁ , F ₂ , F ₃ described in connection with FIG. 6-8, according to the coding format, the downmix section 110 can calculate a two-channel downmix signal L ₁ , L ₂ based on the five-channel audio signal L, LS, LB, TFL, TBL. For example, in the first coding format F ₁ , the first channel L ₁ of the downmix signal is formed as a linear combination (for example, the sum) of the first group 601 of five-channel audio channels L, LS, LB, TFL, TBL, and the second channel L ₂ of the downmix signal is formed in the form of a linear combination (for example, the sum) of the second group 602 channels of the five-channel audio signal L, LS, LB, TFL, TBL. The action performed by the downmix section 110 may be described by expression (1), for example.

For each of the coding formats F ₁ , F ₂ , F ₃ analysis section 120 determines a set of dry upmix coefficients β _L defining a linear mapping of the corresponding downmix signal L ₁ , L ₂ approximating a five-channel audio signal L, LS, LB, TFL , TBL, and calculates the difference between the covariance of the received five-channel audio signal L, LS, LB, TFL, TBL, and the covariance of the five-channel audio signal approximated by corresponding linear mapping of the corresponding downmix signal L ₁ , L ₂ . In the present application, the calculated difference is illustrated by the example of the difference between the covariance matrix of the received five-channel audio signal L, LS, LB, TFL, TBL and the covariance matrix of the five-channel audio signal, approximated by the corresponding linear display of the corresponding downmix signal L ₁ , L ₂ . For each of the coding formats F ₁ , F ₂ , F _{3 ,} analysis section 120 determines a set of wet upmix coefficients γ _L based on the corresponding calculated difference, which, together with dry upmix coefficients β _L , enables parametric reconstruction according to expression (2) of the five-channel audio signal L, LS, LB, TFL, TBL from the downmix signal L ₁ , L ₂ and from the three-channel decorrelated signal determined at the decoder side based on the downmix signal L ₁ , L ₂ . A set of wet upmix coefficients γ _L defines a linear display of the decorrelated signal, in which the covariance matrix of the signal obtained by linear mapping of the decorrelated signal approximates the difference between the covariance matrix of the received five-channel audio signal L, LS, LB, TFL, TBL and the covariance matrix of the five-channel audio signal , approximated by a linear mapping of the downmix signal L ₁ , L ₂ .

The downmix section 110 may, for example, calculate the downmix signal L ₁ , L ₂ in the time domain, i.e., based on the time domain representation of the five-channel audio signal L, LS, LB, TFL, TBL, or in the frequency domain, i.e. e. Based on the frequency domain representation of the five-channel audio signal L, LS, LB, TFL, TBL.

The analysis section 120 may, for example, determine dry upmix coefficients β _L and wet upmix coefficients γ _L based on frequency domain analysis of the five-channel audio signal L, LS, LB, TFL, TBL. The analysis section 120 may, for example, receive the downmix signal L ₁ , L ₂ calculated by the downmix section 110, or may calculate its own version of the downmix signal L ₁ , L ₂ to determine dry upmix coefficients β _L and wet upmix coefficients γ _L .

On FIG. 3 is a generalized block diagram of an audio encoding system 300 comprising an encoding section 100 described in connection with FIG. 1, in accordance with one embodiment. In the present embodiment, audio data, such as that recorded by one or more acoustic transducers 301 or generated by audio production equipment 301, is in the form of an 11.1-channel audio signal described with respect to FIG. 6–8. The quadrature mirror filter (QMF) (or filter bank) analysis section 302 converts the five-channel L, LS, LB, TFL, TBL audio signal, one time segment after another, into a QMF region for processing by the analysis section 100 of the five-channel L, LS, LB, TFL audio signal , TBL, shape of time-frequency tiles. (As will be described in more detail below, the QMF analysis section 302 and the opposite element, the QMF synthesis section 305, are optional). Audio coding system 300 comprises an additional coding section 303 similar to coding section 100 and adapted to encode an additional five-channel audio signal R, RS, RB, TFR, and TBR as an additional two-channel downmix signal R ₁ , R ₂ , and associated additional dry parameters upmix β _R , and additional "wet" upmix parameters γ _R . The QMF analysis section 302 also converts the additional five-channel audio signal R, RS, RB, TFR, and TBR into a QMF region for processing by the additional encoding section 303 .

The control section 304 selects one of the coding formats F ₁ , F ₂ , F ₃ based on the wet and dry upmix coefficients γ _L , γ _R and β _L , β _R determined by the coding section 100 and the additional coding section 303 for corresponding coding formats F ₁ , F ₂ , F ₃ . For example, for each of the coding formats F ₁ , F ₂ , F ₃ control section 304 may calculate the ratio

,

,

where E _wet is the sum of the squares of the wet upmix coefficients γ _L and γ _R , and E _dry is the sum of the squares of the dry upmix coefficients β _L , β _R . The selected coding format may be related to the minimum ratio E of the coding formats F ₁ , F ₂ , F ₃ , i.e., the control section 304 may select the coding format corresponding to the smallest ratio E. The inventors have found that a reduced value of the ratio E can be indicative of the increased fidelity of the 11.1-channel audio signal reconstructed from the corresponding coding format.

In some embodiments, the sum of squares E _{dry of} the dry upmix coefficients β _L , β _R may, for example, include an additional parameter with a value of 1 corresponding to the fact that channel C is transmitted to the decoder side and can be reconstructed without decorrelation, for example , using only a dry upmix factor of 1.

In some embodiments, the control section 304 may select coding formats for the two five-channel audio signals L, LS, LB, TFL, TBL, and R, RS, RB, TFR, TBR independently of each other based on wet and dry upmix coefficients. γ _L , β _L and additional "wet" and "dry" upmix coefficients γ _R , β _R respectively.

The audio encoding system 300 can then output the downmix signal L ₁ , L ₂ and the additional downmix signal R ₁ , R ₂ of the selected coding format, the upmix parameters α from which the dry and wet upmix coefficients β _L can be derived. , γ _L and additional dry and wet upmix coefficients β _R , γ _R associated with the selected coding format, and a signaling S indicating the selected coding format.

In the present embodiment, the control section 304 outputs a downmix signal L ₁ , L ₂ and an additional downmix signal R ₁ , R ₂ of the selected coding format, upmix parameters α from which dry and wet upmix coefficients β can be obtained. _L , γ _L and additional dry and wet upmix coefficients β _R , γ _R associated with the selected coding format, and a signaling S indicating the selected coding format. The downmix signal L ₁ , L ₂ and the additional downmix signal R ₁ , R ₂ are converted back from the QMF domain by a QMF synthesis (or filter bank) section 305 and converted to a modified discrete cosine transform (MDCT) domain by a transform section 306 . The quantization section 307 quantizes the upmix parameters α. For example, you can use uniform quantization with a step size of 0.1 or 0.2 (dimensionless) followed by entropy coding in the form of Huffman coding. A coarser quantization with a step size of 0.2 can be used, for example, to save transmission bandwidth, and a finer quantization with a step size of 0.1 can be used, for example, to increase the reliability of the reconstruction at the decoder side. The C and LFE channels are also converted to the MDCT region by transform section 308 . The transformed MDCT downmix signals and channels, the quantized upmix parameters and signaling are then combined into bitstream B by the multiplexer 309 for transmission to the decoder side. The audio coding system 300 may also comprise a core encoder (not shown in FIG. 3) configured to encode the downmix signal L ₁ , L ₂ , the additional downmix signal R ₁ , R ₂ , and the C and LFE channels using a perceptual audio codec, such as Dolby Digital, MPEG AAC, or derivatives thereof, before transmitting the downmix signals and C and LFE channels to the multiplexer 309. A gain, for example, corresponding to -8.7 dB, can, for example, be applied to the downmix signal L ₁ , L ₂ , an additional downmix signal R ₁ , R ₂ , and channel C until bitstream B is generated _. , L ₂ .

Embodiments may also be envisaged in which the control section 304 obtains only wet and dry upmix coefficients γ _L , γ _R , β _L , β _R for different coding formats F ₁ , F ₂ , F ₃ (or the sum squared wet and dry upmix coefficients for different coding formats) to select the coding format, i.e., the control section 304 need not necessarily receive the downmix signals L ₁ , L ₂ R ₁ , R ₂ for different coding formats. In such embodiments, control section 304 may, for example, control coding sections 100, 303 to transmit downmix signals L ₁ , L ₂ , R ₁ , R ₂ , dry upmix coefficients β _L , β _R and wet coefficients upmix γ _L , γ _R for the selected coding format as an output of the audio coding system 300 or as an input to the multiplexer 309.

If the selected coding format is switched, it is possible, for example, to interpolate between the values of the downmix coefficients used before and after the coding format switch to generate a downmix signal according to equation (1). This is essentially equivalent to interpolating the downmix signals obtained according to the respective sets of downmix coefficient values.

Although in Fig. 3 shows how a downmix signal can be obtained in the QMF domain and then further converted back to the time domain, an alternative encoder can be provided that performs the same functions without using the QMF sections 302, 305, and thereby calculates the downmix signal. mixing directly in the time domain. This is possible in situations where the downmix coefficients are not frequency dependent, as is usually the case. When using an alternative encoder, coding format switching can be processed either by smoothly transitioning between two downmix signals of the respective coding formats, or by interpolation of downmix coefficients (including coefficients that have zero values in one of the formats) producing downmix signals. Such an alternative encoder may have lower delay/latency and/or lower computational complexity.

On FIG. 2 is a generalized block diagram of an encoding section 200 similar to the encoding section 100 described in connection with FIG. 1, in accordance with one embodiment. The encoding section 200 contains a downmix section 210 and an analysis section 220 . As in the encoding section 100 described with respect to FIG. 1, the downmix section 210 calculates a two-channel downmix signal L ₁ , L ₂ based on the five-channel audio signal L, LS, LB, TFL, TBL for each of the coding formats F ₁ , F ₂ , F ₃ , and the analysis section 220 determines the corresponding sets dry upmix coefficients β _L and calculates the differences Δ _L between the covariance matrix of the received five-channel audio signal L, LS, LB, TFL, TBL and the covariance matrices of the five-channel audio signal approximated by the corresponding linear mappings of the respective downmix signals.

Unlike the analysis section 120, in the encoding section 100 described with respect to FIG. 1, analysis section 220 does not calculate wet upmix parameters for all coding formats. Instead, the calculated differences _ΔL are passed to the control section 304 (see FIG. 3) to select the coding format. After selecting the coding format based on the calculated differences _ΔL, the control section 304 may determine wet upmix coefficients (to be included in the upmix parameter set) for the selected coding format. In an alternative embodiment, the control section 304 is responsible for selecting the coding format based on the computed differences Δ _L between the above-described covariance matrices, but instructs the analysis section 220 to calculate the wet upmix coefficients γ _L via signaling backwards; according to this alternative (not shown), the analysis section 220 is capable of outputting both differences and wet upmix coefficients.

In the present exemplary embodiment, a set of wet upmix coefficients is determined such that the covariance matrix of a signal obtained by linear mapping of the decorrelated signal determined by the wet upmix coefficients complements the covariance matrix of a five-channel audio signal approximated by a linear mapping of the downmix signal of the selected coding format . In other words, it is not necessary to define wet upmix parameters in order to achieve full covariance reconstruction when reconstructing the L, LS, LB, TFL, TBL five-channel audio signal at the decoder side. Wet upmix parameters can be determined to increase the fidelity of the reconstructed five-channel audio signal, but if, for example, the number of decorrelators on the decoder side is limited, then wet upmix parameters can be determined so that the five-channel audio covariance matrix L can be reconstructed as much as possible, LS, LB, TFL, TBL.

Embodiments may be envisaged in which audio coding systems similar to the audio coding system 300 described in connection with FIG. 3 comprise one or more coding sections 200 of the type described in connection with FIG. 2.

On FIG. 4 is a flow diagram of an audio encoding method 400 for encoding an M -channel audio signal into a two-channel downmix signal and associated upmix parameters, in accordance with one embodiment. An example of an audio encoding method 400 in the present application is a method implemented by an audio encoding system comprising an encoding section 200 described with respect to FIG. 2.

The audio encoding method 400 includes obtaining 410 a five-channel audio signal L, LS, LB, TFL, TBL; calculation 420 in accordance with the first of the coding formats F ₁ , F ₂ , F ₃ described in connection with FIG. 6-8, two-channel downmix signal L ₁ , L ₂ based on five-channel audio signal L, LS, LB, TFL, TBL; determining 430 a set of dry upmix coefficients β _L according to the coding format; and calculating 440 a difference Δ _L according to the coding format. The audio encoding method 400 includes determining 450 whether differences _ΔL have been computed for each of the coding formats F ₁ , F ₂ , F ₃ . If the difference _ΔL still needs to be determined for at least one coding format, then the audio coding method 400 returns to calculating 420 the downmix signal L ₁ , L ₂ in accordance with the next coding format in turn, as indicated by the letter N in the block diagram.

If the differences _ΔL are computed for each of the coding formats F ₁ , F ₂ , F ₃ , as denoted by Y in the block diagram, method 400 proceeds to select 460 one of the coding formats F ₁ , F ₂ , F ₃ based on the respective calculated differences Δ _L and determining 470 a set of wet upmix coefficients, which, together with dry upmix coefficients β _L of the selected coding format, allow the parametric reconstruction of a five-channel audio signal L, LS, LB, TFL, TBLM in accordance with equation (2 ). The audio encoding method 400 further includes outputting 480 a downmix signal L ₁ , L ₂ of the selected coding format and upmix parameters from which to derive dry and wet upmix coefficients associated with the selected coding format; and output 490 signaling S indicating the selected coding format.

On FIG. 5 is a flowchart of an audio encoding method 500 for encoding an M -channel audio signal into a two-channel downmix signal and associated upmix parameters, in accordance with one embodiment. An example of an audio encoding method 500 in the present application is a method implemented by the audio encoding system 300 described with respect to FIG. 3.

Similar to the audio encoding method 400 described with respect to FIG. 4, the audio encoding method 500 includes deriving 410 a five-channel L, LS, LB, TFL, TBL audio signal; calculating 420 according to the first of the coding formats F ₁ , F ₂ , F ₃ a two-channel downmix signal L ₁ , L ₂ based on the five-channel audio signal L, LS, LB, TFL, TBL; determining 430 a set of dry upmix coefficients β _L in accordance with the coding format; and calculating 440 the difference _ΔL in accordance with the coding format. The audio encoding method 500 further includes determining 560 a set of wet upmix coefficients γ _L , which, in combination with the dry coding format upmix coefficients β _L , allow parametric reconstruction of the M -channel audio signal in accordance with equation (2). Audio encoding method 500 includes determining 550 whether wet and dry upmix coefficients γ _L , β _L have been computed for each of the coding formats F ₁ , F ₂ , F ₃ . If at least one coding format still needs to calculate wet and dry upmix coefficients γ _L , β _L , then the audio encoding method 500 returns to the calculation 420 of the downmix signal L ₁ , L ₂ in accordance with the following in turn coding format, as indicated by the letter N in the block diagram.

If the wet and dry upmix coefficients γ _L , β _L are computed for each of the coding formats F ₁ , F ₂ , F ₃ , as indicated by Y in the block diagram, the audio encoding method 500 proceeds to select 570 one of coding formats F ₁ , F ₂ , F ₃ based on the respective calculated wet and dry upmix coefficients γ _L , β _L ; outputting 480 the downmix signal L ₁ , L ₂ of the selected coding format and upmix parameters from which dry and wet upmix coefficients β _L , γ _L associated with the selected coding format can be derived; and a signaling output 490 indicating the selected coding format.

On FIG. 9 is a generalized block diagram of a decoding section 900 for reconstructing an M -channel audio signal based on a two-channel downmix signal and associated upmix parameters α _L , in accordance with one embodiment.

In the present embodiment, an example of the downmix signal is the L ₁ , L ₂ downmix signal that is output by the encoding section 100 described with respect to FIG. 1. In the present embodiment, the dry and wet upmix coefficients β _L , γ _L , which are output by the encoding section 100 and which are adapted for parametric reconstruction of the five-channel audio signal L, LS, LB, TFL, TBL, are obtained from the upmix parameters α _L . However, embodiments may also be envisaged in which the upmix parameters α _L are adapted for parametric reconstruction of an M -channel audio signal, where M = 4 or M ≥ 6.

Section 900 decoding contains section 910 predecrelation, section 920 decorrelation and section 930 mixing. The pre-decorrelation section 910 determines a set of pre-decorrelation coefficients based on the selected coding format used on the encoder side to encode the L, LS, LB, TFL, TBL five-channel audio signal. As described below with respect to FIG. 10, the selected coding format may be indicated by signaling from the encoder side. The pre-decrelation section 910 calculates the input decorrelation signal D ₁ , D ₂ , D ₃ as a linear mapping of the down-mix signal L ₁ , L ₂ , where a set of pre-decoration coefficients is applied to the down-mix signal L ₁ , L ₂ .

The decorrelation section 920 generates a decorrelated signal based on the decorrelation input D ₁ , D ₂ , D ₃ . Here, an example of a decorrelated signal is three channels, each generated by processing one of the decorrelation input channels in the decorrelator 921-923 of the decorrelation section 920, for example, including applying linear filters to the corresponding decorrelation input channels D ₁ , D ₂ , D ₃ .

, соответствующего подлежащему реконструкции пятиканальному аудиосигналу L, LS, LB, TFL, TBL.The mixing section 930 determines a set of wet and dry upmix coefficients β _L , γ _L based on the obtained upmix parameters α _L and the selected coding format used on the encoder side to encode the five-channel audio signal L, LS, LB, TFL, TBL . The mixing section 930 performs parametric reconstruction of the five-channel audio signal L, LS, LB, TFL, TBL in accordance with equation (2), i.e., calculates the dry upmix signal as a linear display of the downmix signal L ₁ , L ₂ , and to the downmix signal L ₁ , L ₂ apply a set of "dry" upmix coefficients β _L ; calculates the wet upmix signal as a linear mapping of the decorrelated signal, wherein a set of wet upmix coefficients γ _L is applied to the decorrelated signal; and combines dry and wet upmix signals to produce a multi-dimensional reconstructed audio signal

corresponding to the five-channel audio signal to be reconstructed L, LS, LB, TFL, TBL.

In some embodiments, the obtained upmix parameters α _L may include the wet and dry upmix coefficients β _L , γ _L themselves, or they may follow a more compact form including fewer parameters than the number of wet and dry upmix coefficients β _L , γ _L , from which wet and dry upmix coefficients β _L , γ _L can be obtained on the decoder side, knowing the compact form used.

On FIG. 11 shows the operation of the mixing section 930 described in connection with FIG. 9, in one example scenario in which the downmix signal L ₁ , L ₂ represents a five-channel audio signal L, LS, LB, TFL, TBL according to the first coding format F ₁ described with respect to FIG. 6. It should be understood that in example scenarios in which the downmix signal L ₁ , L ₂ represents a five-channel audio signal L, LS, LB, TFL, TBL in accordance with any of the second and third coding formats F ₂ , F ₃ , section operation 930 mixing may be similar. In particular, the mixing section 930 may temporarily activate additional instances of upmix sections and merging sections, which will be described later, and provide a smooth transition between the two coding formats, which may require the computed downmix signals to be available simultaneously.

In the present scenario example, the first channel L ₁ of the downmix signal represents three channels L, LS, LB, and the second channel L ₂ of the downmix signal represents two channels TFL, TBL. The pre-decorrelation section 910 determines pre-decrelation coefficients to generate two channels of a decorrelated signal based on the first channel L ₁ of the downmix signal and to generate one channel of the decorrelated signal based on the second channel L ₂ of the downmix signal.

каналов L, LS, LB.The first dry upmix section 931 provides a three-channel dry upmix signal X ₁ as a linear mapping of the first channel L ₁ of the downmix signal, wherein a subset of the dry upmix coefficients that can be obtained from the obtained upmix parameters α _L are applied to the first channel L ₁ of the downmix signal. The first wet upmix section 932 provides a three-channel wet upmix signal Y ₁ as a linear mapping of the two channels of the decorrelated signal, with a subset of wet upmix coefficients that can be obtained from the obtained upmix parameters α _L applied to the two channels decorrelated signal. The first merging section 933 combines the first dry upmix signal X ₁ and the first wet upmix signal Y ₁ into reconstructed versions.

channels L, LS, LB.

каналов TFL, TBL.Similarly, the second dry upmix section 934 provides a two-channel dry upmix signal X ₂ as a line mapping of the second channel L ₂ of the downmix signal, and the second wet upmix section 935 provides a two-channel wet upmix signal Y ₂ in in the form of a linear combination of one channel of the decorrelated signal. The second merging section 936 combines the second dry upmix signal X ₂ and the second wet upmix signal Y ₂ into reconstructed versions.

channels TFL, TBL.

On FIG. 10 is a generalized block diagram of an audio decoding system 1000 comprising a decoding section 900 described in connection with FIG. 9 in accordance with one embodiment. An acquisition section 1001, for example including a demultiplexer, obtains the B bitstream transmitted from the audio coding system 300 described with respect to FIG. 3 and extracts the downmix signal L ₁ , L ₂ , the additional downmix signal R ₁ , R ₂ , and the upmix parameters α as well as the C and LFE channels from bitstream B. The upmix parameters α may, for example, comprise the first and the second subsets α _L and α _R associated with the left side and right side, respectively, of the 11.1-channel audio signal L, LS, LB, TFL, TBL, R, RS, RB, TFR, TBR, C, LFE to be reconstructed.

If the downmix signal L ₁ , L ₂ , the additional downmix signal R ₁ , R ₂ , and/or the C and LFE channels are encoded in the B bitstream using a perceptual audio codec such as Dolby Digital, MPEG AAC or derivatives thereof, decoding system 1000 the audio may comprise a basic decoder (not shown in FIG. 10) configured to decode the respective signals and channels when extracted from bitstream B.

The conversion section 1002 converts the L ₁ , L ₂ downmix signal by performing inverse MDCT, and the QMF analysis section 1003 converts the L ₁ , L ₂ downmix signal to a QMF domain for processing by the L ₁ , L ₂ downmix signal decoding section 900, form time-frequency tiles. The dequantization section 1004 dequantizes the first subset of the upmixing parameters α _L , for example, from the entropy encoding format, before passing it to the decoding section 900 . As described with respect to FIG. 3, quantization may be performed using one of two different step sizes, such as 0.1 or 0.2. The actual step size used may be predetermined or may be signaled to the audio decoding system 1000 by the encoder, such as via bitstream B.

дополнительного пятиканального аудиосигнала R, RS, RB, TFR, TBR на основе дополнительного сигнала понижающего микширования R₁, R₂ и второго поднабора α_R параметров повышающего микширования.In the present embodiment, the audio decoding system 1000 includes an additional decoding section 1005 similar to the decoding section 900 . The additional decoding section 1005 is configured to obtain an additional two-channel downmix signal R ₁ , R ₂ described in connection with FIG. 3 and a second subset α _R of the upmix and provide reverse engineer parameters

an additional five-channel audio signal R, RS, RB, TFR, TBR based on the additional downmix signal R ₁ , R ₂ and the second subset α _R of the upmix parameters.

The conversion section 1006 converts the additional downmix signal R ₁ , R ₂ by performing inverse MDCT, and the QMF analysis section 1007 converts the additional downmix signal R ₁ , R ₂ into a QMF region for processing by the additional section 1005 decoding the additional downmix signal R ₁ , R ₂ , the shape of the time-frequency tiles. The dequantizing section 1008 dequantizes the second subset of the upmixing parameters α _R , for example, from the entropy encoding format, before passing them to the further decoding section 1005 .

In embodiments in which a gain factor has been applied to the downmix signal L ₁ , L ₂ , the additional downmix signal R ₁ , R _{2 ,} and channel C on the encoder side, an appropriate gain, for example corresponding to 8.7 dB, can be applied to these signals in audio decoding system 1000 for gain compensation.

The control section 1009 receives the signaling S indicating the format selected from the coding formats F ₁ , F ₂ , F ₃ used on the encoder side to encode the 11.1-channel audio signal into the downmix signal L ₁ , L ₂ , as well as the additional downmix signal R ₁ , R ₂ and associated upmix parameters α. The control section 1009 controls the decoding section 900 (for example, its included pre-decorrelation section 910 and mixing section 920) and the additional decoding section (1005), and performs parametric reconstruction in accordance with the specified encoding format.

In the present embodiment, the reconstructed versions of the five-channel audio signal L, LS, LB, TFL, TBL and the additional five-channel audio signal R, RS, RB, TFL, TBL output by the decoding section 900 and the additional decoding section 1005, respectively, are converted back from the QMF region by the section 1011 QMF synthesis, and then output together with the C and LFE channels as the output of the audio decoding system 1000 for playback on the multi-column system 1012. The conversion section 1010 converts the C and LFE channels into the time domain by performing inverse MDCT, and then these channels are included in the output of the audio decoding system 1000.

Channels C and LFE may, for example, be extracted from bitstream B in a discretely encoded form, and the audio decoding system 1000 may, for example, comprise single-channel decoding sections (not shown in FIG. 10) configured to decode the respective discretely encoded channels. . The single-channel decoding sections may, for example, include core decoders for decoding audio data encoded with a perceptual audio codec such as Dolby Digital, MPEG AAC, or derivatives thereof.

In the present exemplary embodiment, the pre-decrelation coefficients are determined by the pre-decrelation section 910 such that, in each of the coding formats F ₁ , F ₂ , F ₃ , each of the decorrelation input signal channels D ₁ , D ₂ , D ₃ coincides with the downmix signal channel L ₁ , L ₂ in accordance with table 1.

As can be seen in Table 1, the TBL channel contributes via the downmix signal L ₁ , L ₂ to the third channel D3 of the decorrelation input signal in all three coding formats F ₁ , F ₂ , F ₃ , while each of the channel pairs LS, LB and TFL, TBL contributes via the downmix signal L ₁ , L ₂ to the third decorrelation signal channel D3 in at least two coding formats, respectively.

Table 1 shows that each of the channels L and TFL contributes via the downmix signal L ₁ , L ₂ to the first channel D1 of the input decorrelation signal in two coding formats, respectively, and the pair of channels LS, LB contributes via the downmix signal L ₁ , L ₂ into the first channel D1 of the decorrelation input signal in at least two coding formats.

Table 1 also shows that the three channels LS, LB, TBL contribute via the downmix signal L ₁ , L ₂ to the second channel D2 of the decorrelation input signal in both the second and third coding formats F ₃ , F ₃ , while the channel pair LS, LB contributes via the downmix signal L ₁ , L ₂ to the second channel D2 of the decorrelation input signal in all three coding formats F ₁ , F ₂ , F ₃ .

Switching between different coding formats changes the input signal to decorrelators 921-923. In the present embodiment, at least some portions of the decorrelation inputs D1, D2, D3 will be retained when switching, i.e., at least one channel of the five-channel audio signal L, LS, LB, TFL, TBL will remain in each channel of the decorrelation input D1, D2, D3 at any switch between the two coding formats F ₁ , F ₂ , F ₃ , which will allow a smoother transition between coding formats when perceived by the listener when playing the reconstructed M -channel audio signal.

The inventors have found that since a decorrelated signal can be generated based on a portion of the downmix signal L ₁ , L ₂ corresponding to several time intervals during which coding format switching can occur, it is potentially possible for audible distortions to be generated in the decorrelated signal as a result of format switching coding. Even if the wet and dry upmix coefficients β _L , γ _L are interpolated in response to the transition between coding formats, the distortions that have occurred in the decorrelated signal may still be present in the reconstructed five-channel audio signal L, LS, LB, TFL, TBL. Providing decorrelation input D1, D2, D3 according to Table 1 can suppress audible distortion in the decorrelated signal caused by coding format switching, and can improve the playback quality of the reconstructed L, LS, LB, TFL, TBL five-channel audio signal.

Although Table 1 is expressed in terms of the coding formats F ₁ , F ₂ , F ₃ for which the downmix signal channels L ₁ , L ₂ are generated as the sums of the first and second channel groups, respectively, the same values of the pre-decoration coefficients can, for example, be used, when the downmix signal channels are formed as linear combinations of the first and second channel groups, respectively, so that the decorrelation signal input channels D1, D2, D3 coincide with the downmix signal channels L ₁ , L ₂ in accordance with Table 1. It should be understood that the quality reproduction of the reconstructed five-channel audio signal in this manner can also be improved if the channels of the downmix signal are formed as linear combinations of the first and second channel groups, respectively.

To further improve the reproduction quality of the reconstructed five-channel audio signal, for example, interpolation of the values of pre-decoration coefficients can be performed in response to switching the coding format. In the first coding format F _{1 ,} the decorrelation input signal D1, D2, D3 can be defined as follows:

while in the second coding format F ₂ the decorrelation input signal D1, D2, D3 can be defined as follows:

In response to switching from the first coding format F ₁ to the second coding format F ₂ , continuous or linear interpolation can be performed between the pre-decoration matrix in equation (3) and the pre-decoration matrix in equation (4), for example.

The downmix signal L ₁ , L ₂ in equations (3) and (4) may, for example, be present in the QMF region, and when switching between coding formats, the downmix coefficients applied at the encoder side to calculate the downmix signal L ₁ , L ₂ according to equation (1) can be interpolated over 32 QMF intervals, for example. The interpolation of the pre-decoration coefficients (or matrices) may, for example, be synchronized with the interpolation of the downmix coefficients, for example, it may be performed over the same 32 QMF intervals. The interpolation of the pre-decoration coefficients may be, for example, a wideband interpolation, such as used for all frequency bands decoded by the audio decoding system 1000.

The dry and wet upmix coefficients β _L , γ _L can also be interpolated. The interpolations of the dry and wet upmix coefficients β _L , γ _L can, for example, be controlled by signaling S on the encoder side to improve transition processing. When switching the coding format, the interpolation scheme selected at the encoder side to interpolate dry and wet upmix coefficients β _L , γ _L at the decoder side may be, for example, an interpolation scheme suitable for switching the coding format, which may be different from the interpolation schemes used for dry and wet upmix coefficients β _L , γ _L in the absence of coding format switching.

In some embodiments, at least one interpolation scheme different from additional decoding section 1005 may be applied in decoding section 900.

On FIG. 12 is a flowchart of an audio decoding method 1200 for reconstructing an M -channel audio signal based on a two-channel downmix signal and associated upmix parameters, in accordance with one embodiment. An example of a decoding method 1200 in the present application is a decoding method that can be implemented by the audio decoding system 1000 described with respect to FIG. 10.

An audio decoding method 1200 includes deriving 1201 a two-channel downmix signal L ₁ , L ₂ and upmix parameters α _L to parametrically reconstruct the five-channel audio signal L, LS, LB, TFL, TBL described with respect to FIG. 6-8, based on the downmix signal L ₁ , L ₂ ; receiving 1202 signaling S indicating a format selected from the coding formats F ₁ , F ₂ , F ₃ described in connection with FIG. 6–8; and determining 1203 a set of pre-decoration coefficients based on the specified coding format.

Audio decoding method 1200 includes detecting 1204 whether a switch is in progress from one encoding format to another. If the switch is not detected, as indicated by N in the block diagram, then the next step is to compute 1205 the decorrelation input D ₁ , D ₂ , D ₃ as a linear mapping of the downmix signal L ₁ , L ₂ , with the set of predecorrelation coefficients. On the other hand, if a coding format switch is detected, as indicated by the letter Y in the flowchart, then instead the next step is to interpolate 1206 in the form of a gradual transition from pre-decoration coefficient values of one coding format to pre-decoration coefficient values of another coding format, and then calculate 1205 decorrelation input signal D ₁ , D ₂ , D ₃ using the interpolated values of the pre-decoration coefficients.

An audio decoding method 1200 includes generating 1207 a decorrelated signal based on the decorrelation input signal D ₁ , D ₂ , D ₃ ; and determining 1208 a set of wet and dry upmix coefficients β _L , γ _L based on the obtained upmix parameters and the specified coding format.

, соответствующего подлежащему реконструкции пятиканальному аудиосигналу.If no coding format switch is detected, as indicated by branch N from condition check block 1209, then method 1200 proceeds to compute 1210 the dry upmix signal as a linear mapping of the downmix signal, with _the set _" dry" upmix coefficients β _L ; and calculating 1211 the wet upmix signal as a linear mapping of the de-correlated signal, wherein a set of wet up-mix coefficients γ _L is applied to the de-correlated signal. On the other hand, if a switch from one coding format to another is detected, as indicated in the flow diagram by the Y branch from the condition check block 1209, then the method instead proceeds to perform 1212 interpolation from the dry and wet upmix coefficient values ( including coefficients of zero) applicable to one coding format to dry and wet upmix coefficients (including coefficients of zero) applicable to another coding format; calculating 1210 the dry upmix signal as a linear mapping of the downmix signal L ₁ , L ₂ , wherein an interpolated set of dry upmix coefficients is applied to the downmix signal L ₁ , L ₂ ; and calculating 1211 the wet upmix signal as a linear mapping of the decorrelated signal, wherein an interpolated set of wet upmix coefficients is applied to the decorrelated signal. The method also includes combining 1213 dry and wet upmix signals to obtain a multi-dimensional reconstructed audio signal.

corresponding to the five-channel audio signal to be reconstructed.

On FIG. 13 is a generalized block diagram of a decoding section 1300 for reconstructing a 13.1-channel audio signal based on a 5.1-channel audio signal and associated upmix parameters α, in accordance with one embodiment.

In the present embodiment, 13.1-channel audio examples are LW (Wide Left), LSCRN (Screen Left), TFL (Front Top Left), LS (Surround Left), LB (Surround Left), TBL (Surround Top Left), RW (Wide Right), RSCRN (Screen Right), TFR (Top Front Right), RS (Right Side), RB (Right Surround), TBR (Top Surround Right), C (Center), and LFE (Low Frequency Effects Channel). This 5.1-channel signal includes a downmix signal L ₁ , L ₂ for which the first channel L ₁ corresponds to a linear combination of channels LW, LSCRN, TFL and for which the second channel L ₂ corresponds to a linear combination of channels LS, LB, TBL; and an additional downmix signal R ₁ , R ₂ , for which the first channel R ₁ corresponds to a linear combination of channels RW, RSCRN, TFR and for which the second channel R ₂ corresponds to a linear combination of channels RS, RB, TBR; and channels C and LFE.

13.1-канального аудиосигнала может обеспечиваться в качестве выхода секции 1310 декодирования.The first upmix section 1310 reconstructs the channels LW, LSCRN and TFL based on the first channel L ₁ of the downmix signal under the control of at least some of the upmix parameters α; the second upmix section 1320 reconstructs the channels LS, LB, TBL based on the second channel L ₂ of the downmix signal under the control of at least some of the upmix parameters α; the third upmix section 1330 reconstructs the channels RW, RSCRN, TFR based on the first channel R ₁ of the additional downmix signal under the control of at least some of the upmix parameters α; and the fourth upmix section 1340 reconstructs the channels RS, RB, TBR based on the second channel R ₂ of the downmix signal under the control of at least some of the upmix parameters α. Remodeled variant

A 13.1-channel audio signal may be provided as an output of the decoding section 1310.

In an exemplary embodiment, the audio decoding system 1000 described in connection with FIG. 10 may include a decoding section 1300 in addition to the decoding sections 900 and 1005, or may at least be capable of reconstructing a 13.1 channel signal in a manner similar to that performed in the decoding section 1300. Signaling S extracted from bitstream B may, for example, indicate whether the received 5.1-channel audio signal L ₁ , L ₂ , R ₁ , R ₂ , C, LFE and associated upmix parameters represents a 11.1-channel signal as described in relation to Fig. 10, or it represents a 13.1 channel audio signal as described with respect to FIG. 13.

The control section 1009 may determine whether the received signaling S indicates an 11.1-channel or 13.1-channel configuration, and may control other sections of the audio decoding system 1000 when performing parametric reconstruction of either the 11.1-channel audio signal as described with respect to FIG. 10 or 13.1 channel audio as described with respect to FIG. 13. For a 13.1-channel configuration, for example, a single coding format can be used instead of the two or three coding formats used for a 11.1-channel configuration. If the S signaling indicates a 13.1-channel configuration, the coding format may thus be implicitly specified, and there may be no need for S signaling to explicitly indicate the selected coding format.

It should be understood that while the embodiments described with respect to FIGS. 1-5 are set forth in terms of the 11.1-channel audio signal described with respect to FIG. 6-8, coding systems may be provided that may include any number of coding sections and that may be configured to encode any number of M -channel audio signals, where M ≥ 4. Similarly, it should be understood that while the embodiments described with respect to FIG. . 9-12 are set forth in terms of the 11.1-channel audio signal described with respect to FIG. 6-8, decoding systems may be provided which may include any number of decoding sections and which may be capable of reconstructing any number of M -channel audio signals, where M ≥ 4.

In some embodiments, the encoder side can choose between three coding formats F ₁ , F ₂ , F ₃ . In other embodiments, the encoder side can only choose between two coding formats, such as the first and second coding formats F ₁ , F ₂ .

On FIG. 14 is a generalized block diagram of an encoding section 1400 for encoding an M -channel audio signal as a two-channel downmix signal and associated dry and wet upmix coefficients, in accordance with one embodiment. The coding section 1400 may be in an audio coding system of the type shown in FIG. 3. More specifically, it may be located in the space occupied by the encoding section 100. As will become apparent from the description of the internal workings of the shown components, the encoding section 1400 can operate in two different encoding formats; however, it is possible to implement similar coding sections within the scope of the invention that are capable of operating in three or more coding formats.

The encoding section 1400 contains a downmix section 1410 and an analysis section 1420. For at least one selected format (see description of control section 1430 of encoding section 1400 below) from encoding formats F ₁ , F ₂ , which may be one of the formats described with respect to FIG. 6-7, or any of the other formats, the downmix section 1410 calculates, according to the coding format, a two-channel downmix signal L ₁ , L ₂ based on the five-channel audio signal L, LS, LB, TFL, TBL. For example, in the first coding format F ₁ , the first channel L ₁ of the downmix signal is formed as a linear combination (for example, the sum) of the first group of channels of the five-channel audio signal L, LS, LB, TFL, TBL, and the second channel L ₂ of the downmix signal is formed in in the form of a linear combination (for example, sum) of the second group of channels of a five-channel audio signal L, LS, LB, TFL, TBL. The action performed by the downmix section 1410 can be described by expression (1), for example.

For said at least one selected format from the coding formats F ₁ , F _{2 ,} analysis section 1420 determines a set of dry upmix coefficients β _L defining a linear mapping of the corresponding downmix signal L ₁ , L ₂ approximating a five-channel audio signal L, LS, LB, TFL, TBL. For each of the coding formats F ₁ , F ₂ , analysis section 1420 further determines a set of wet upmix coefficients γ _L based on the corresponding computed difference, which, in combination with dry upmix coefficients β _L , allow parametric reconstruction according to equation (2 ) five-channel audio signal L, LS, LB, TFL, TBL from the downmix signal L ₁ , L ₂ and from the three-channel decorrelated signal determined at the decoder side based on the downmix signal L ₁ , L ₂ . A set of wet upmix coefficients γ _L defines a linear display of the decorrelated signal, in which the covariance matrix of the signal obtained by linear mapping of the decorrelated signal approximates the difference between the covariance matrix of the received five-channel audio signal L, LS, LB, TFL, TBL and the covariance matrix of the five-channel audio signal , approximated by a linear mapping of the downmix signal L ₁ , L ₂ .

The downmix section 1410 may, for example, calculate the downmix signal L ₁ , L ₂ in the time domain, i.e., based on the time domain representation of the five-channel audio signal L, LS, LB, TFL, TBL, or in the frequency domain, i.e. e. Based on the frequency domain representation of the five-channel audio signal L, LS, LB, TFL, TBL. Calculating L ₁ , L ₂ in the time domain is possible, at least if the choice of coding format is not frequency selective, and therefore applicable to all frequency components of the M -channel audio signal; this is currently the preferred option.

The analysis section 1420 may, for example, determine dry upmix coefficients β _L and wet upmix coefficients γ _L based on frequency domain analysis of the five-channel audio signal L, LS, LB, TFL, TBL. Frequency domain analysis can be performed on a cropped portion of an M -channel audio signal. Cropping can, for example, use split rectangular or overlapping triangular windows. The analysis section 1420 may, for example, receive the _downmix signal L ₁ , L ₂ calculated by the downmix section 1410 (not shown _in FIG. » upmix coefficients β _L and wet upmix coefficients γ _L .

The encoding section 1400 further comprises a control section 1430 which is responsible for selecting the format currently in use. It is not necessary for the control section 1430 to use a certain criterion or a certain basis when deciding on the selection of an encoding format. The signaling value S generated by the control section 1430 indicates the result of the decision made by the control section 1430 for the currently considered portion (eg, time slot) of the M -channel audio signal. The S signaling may be included in the B bitstream generated by the coding system 300, which includes the coding section 1400, to simplify the reconstruction of the encoded audio signal. In addition, signaling S is transmitted to each of the downmix section 1410 and the analysis section 1420 to inform these sections of the coding format being used. Like the analysis section 1420, the control section 1430 may consider cropped portions of the M -channel signal. For completeness, it should be noted that the downmix section 1410 may operate with a delay of 1 or 2 frames, and possibly with an additional advance relative to the control section 1430. Signaling S may also optionally contain information related to ensuring a smooth transition of the downmix signal generated by the downmix section 1410 and/or information related to decoder-side interpolation of the sample values of dry and wet coefficients of the upmix that the section provides. 1420 analysis, to ensure synchronism on the sub-frame timeline.

As an optional component, the encoding section 1400 may include a stabilizer 1440 located immediately after the control section 1430 and acting on its output just before it is processed by other components. Based on this output, stabilizer 1440 provides additional information S to downstream components. Stabilizer 1440 may achieve the desired goal of not changing the coding format too often. To this end, stabilizer 1440 may take into account the number of coding format selections over the elapsed time slots of the M -channel audio signal and ensure that the selected coding format is retained for at least a given number of time slots. In an alternative embodiment, the stabilizer may apply an averaging filter to a number of coding format choices made (eg, represented by a discrete variable), which may create a smoothing effect. In yet another embodiment, stabilizer 1440 may comprise a state machine configured to transmit side information S for all slots in the moving time window if the state machine determines that the coding format selection made by control section 1430 has remained unchanged throughout the moving time window. . The moving time window may correspond to a buffer storing the results of the selection of coding formats for a certain number of elapsed time intervals. One skilled in the art will readily appreciate that such stabilization functions may need to be accompanied by an increase in operational delay between the stabilizer 1440 and at least the downmix section 1410 and the analysis section 1420. The delay can be implemented using M -channel audio buffering sections.

As described above, FIG. 14 is a partial view of the coding system shown in FIG. 3. Although the components shown in FIG. 14 relate only to the processing of the left-side channels L, LS, LB, TFL, TBL, the coding system also processes at least the right-side channels R, RS, RB, TFR, TBR. For example, an additional instance (eg, a functionally equivalent copy) of the coding section 1400 may operate in parallel to encode a right side signal, including the indicated R, RS, RB, TFR, TBR channels. Although the left and right channels participate in two separate downmix signals (or at least separate channel groups of the common downmix signal), it is preferable to use a common coding format for all channels. In other words, the control section 1430 in the left-side coding section 1400 may be responsible for deciding on a common coding format for use by both left-side and right-side channels; thus, it is preferable that the control section 1430 also has access to the right-side channels R, RS, RB, TFR, TBR or scores obtained from these signals, for example, covariance, downmix signal, etc., and can take into account them when choosing the encoding format to use. Then, signaling S is transmitted not only to the downmix section 1410 and the analysis section 1420 of the control section 1430 (left-hand), but also to the equivalent sections of the right-hand coding section (not shown). In an alternative embodiment, the goal of using a common coding format for all channels can be achieved if the control section 1430 itself is common to the left and right instances of the control section 1400. In a circuit of the type shown in Fig. 3, the encoding section 1430 can be performed outside of both the encoding section 100 and the additional encoding section 303 responsible for the left-side and right-side channels, respectively, and can receive all the left-side and right-side channels L, LS, LB, TFL, TBL, R, RS. , RB, TFR, TBR, and output signaling S, which indicates the selection of the coding format, and is transmitted to at least the coding section 100 and the additional coding section 303.

On FIG. 15 schematically shows a possible implementation of a downmix section 1410 configured to switch according to signaling S between two given coding formats F ₁ , F ₂ and provide a smooth transition between them. The downmix section 1410 comprises two downmix subsections 1411, 1412 configured to receive an M -channel audio signal and output a two-channel downmix signal. The two downmix subsections 1411, 1412 may be functionally equivalent copies of the same circuit, albeit with different downmix parameters (eg, coefficient values for deriving an L ₁ , L ₂ downmix signal based on an M -channel audio signal). In normal operation, the two downmix subsections 1411, 1412 together provide one downmix signal L ₁ (F ₁ ), L ₂ (F ₁ ) according to the first coding format F ₁ and/or one downmix signal L ₁ (F ₂ ) , L ₂ (F ₂ ) in accordance with the second coding format F ₂ . After the downmix subsections 1411, 1412, the first downmix interpolation section 1413 and the second downmix interpolation section 1414 are located. The first downmix interpolation section 1413 is configured to interpolate, including providing a smooth transition, the first channel L ₁ of the downmix signal, and the second downmix interpolation section 1414 is configured to interpolate, including ensuring a smooth transition, the second signal channel L ₂ downmix. The first downmix interpolation section 1413 may operate in at least the following states:

a) only the first coding format (L ₁ = L ₁ (F ₁ )) can be used with stable operation in the first coding format;

b) only the second coding format (L ₁ = L ₁ (F ₂ )) can be used with stable operation in the second coding format; And

c) combining downmix channels according to both coding formats (L ₁ = α ₁ L ₁ (F ₁ ) + α ₂ L ₁ (F ₂ ), where 0 < α ₁ < 1 and 0 < α ₂ < 1) can be to be used when switching from the first to the second encoding format or vice versa.

It may be necessary for the combining state (c) that downmix signals from both the first and second downmix subsections 1411, 1412 are available. it is possible to transition in small substeps or even with a pseudo-continuous smooth transition. This has the advantage that the smooth transition becomes less noticeable. For example, in the interpolator scheme, where α ₁ + α ₂ = 1, a five-step smooth transition is possible if the following values (α ₁ , α ₂ ) are defined: (0.2; 0.8); (0.4; 0.6); (0.6; 0.4); (0.8; 0.2). The second downmix interpolation section 1414 may have identical or similar capabilities.

In one version of the above embodiment, the downmix section 1410, indicated by the dotted line in FIG. 15, signaling S may also be applied to the first and second downmix subsections 1411, 1412. As described above, the generation of the downmix signal associated with the unselected coding format may be further suppressed. This can reduce the average computing load.

As an addition or alternative to this version, a smooth transition between the downmix signals of two different coding formats can be done by smooth transition of the downmix coefficients. The first downmix sub-section 1411 can then be fed with downmix coefficients generated by a coefficient interpolator (not shown), which stores downmix coefficient preset values for use in the available coding formats F ₁ , F ₂ , and receives signaling S as input. In such a configuration the entire second downmix subsection 1412 and the first and second interpolation subsections 1413, 1414 can be removed or completely deactivated.

The S signaling that the downmix section 1410 receives is transmitted to at least the downmix interpolation sections 1413, 1414, but not necessarily to the downmix subsections 1411, 1412. It is necessary to send the S signaling to the downmix subsections 1411, 1412 if intermittent operation is desired, i.e. if the amount of redundant downmix is to be reduced outside of coding format switching. The signaling may be low-level instructions, such as those related to different modes of operation of the downmix interpolation sections 1413, 1414, or it may be associated with high-level instructions, such as the order in which a given glide program is executed (for example, a sequence of modes of operation where each has a given duration) at a given starting point.

On FIG. 16 shows a possible implementation of an analysis section 1420 operable alternately according to signaling S in two given coding formats F ₁ , F ₂ . Analysis section 1420 includes two analysis subsections 1421, 1422 configured to receive an M -channel audio signal and output dry and wet upmix coefficients. These two analysis subsections 1421, 1422 may be functionally equivalent copies of the same schema. In normal operation, the two analysis subsections 1421, 1422 together provide one set of dry and wet upmix coefficients β _L (F ₁ ), γ _L (F ₁ ) according to the first coding format F ₁ and/or one set of " dry and wet upmix coefficients β _L (F ₂ ), γ _L (F ₂ ) according to the second coding format F ₂ .

As described above for analysis section 1420 as a whole, the current downmix signal may be obtained from downmix section 1410, or a duplicate of this signal may be generated in analysis section 1420. More specifically, the first analysis subsection 1421 may either receive the downmix signal L ₁ (F ₁ ), L ₂ (F ₁ ) according to the first F ₁ coding format from the first downmix subsection 1411 of the downmix section 1410, or may generate its own duplicate. Similarly, the second analysis subsection 1422 may either receive the downmix signal L ₁ (F ₂ ), L ₂ (F ₂ ) according to the second F ₂ coding format from the second downmix subsection 1412 or may generate its own duplicate of this signal.

Following the analysis sections 1421, 1422 are a dry upmix coefficient selector 1423 and a wet upmix coefficient selector 1424. The dry upmix coefficient selector 1423 is configured to redirect the dry upmix coefficient set β _L from either the first or second analysis subsection 1421, 1422, and the wet upmix coefficient selector 1424 is configured to redirect the wet upmix coefficient set. » upmix coefficients γ _L from either the first or second analysis subsection 1421, 1422. The dry upmix coefficient selector 1423 may operate in at least the states (a) and (b) described above with respect to the first downmix interpolation section 1413. However, if the coding system shown in FIG. 3, part of which is described in this part of the section, is configured to work with a decoding system such as that shown in FIG. 9 that performs parametric reconstruction based on the interpolated samples of the obtained upmix coefficients, it is not necessary to adjust the combining state such as (c) defined for the downmix interpolation sections 1413, 1414. Wet upmix coefficient selector 1424 may have similar capabilities.

The signaling S that the analysis section 1420 receives is transmitted to at least the wet and dry upmix coefficient selectors 1423, 1424. It is not necessary for the analysis subsections 1421, 1422 to receive signaling, although it is advantageous to avoid unnecessary calculations of upmix coefficients outside of transitions. The signaling may be low-level commands, such as those related to the different modes of operation of the dry and wet upmix ratio selectors 1423, 1424, or it may be associated with high-level commands, such as the order in which one coding format should be changed to another in a particular time interval. As described above, this preferably includes the operation of providing a smooth transition, but may be as simple as determining the values of the upmix coefficients for an appropriate time point, or determining these values to apply at an appropriate point in time.

Next, a method 1700 will be described, which is a variant of a method for encoding an M -channel audio signal as a two-channel downmix signal according to one embodiment, which is schematically shown as a block diagram in FIG. 17. The exemplary method described in this part of the section may be implemented by an audio encoding system comprising an encoding section 1400 described above with respect to FIG. 14–16.

Audio encoding method 1700 includes: deriving 1710 M -channel L, LS, LB, TFL, TBL audio signal; selecting 1720 one of at least two coding formats F ₁ , F ₂ , F ₃ described in connection with FIG. 6–8; calculating 1730 for the selected coding format of the two-channel downmix signal L ₁ , L ₂ based on the M -channel audio signal L, LS, LB, TFL, TBL; outputting 1740 the downmix signal L ₁ , L ₂ of the selected coding format and side information α allowing parametric reconstruction of the M -channel audio signal based on the downmix signal; and output 1750 signaling S indicating the selected coding format. The method is repeated, for example, for each time interval of the M -channel audio signal. If the output signal of the selection operation 1720 has a different coding format than the one immediately selected previously, then the downmix signal is replaced at an appropriate time by a smooth transition between the downmix signals according to the previous and current coding formats. As already described above, performing a smooth transition of side information, which can be subjected to internal interpolation at the decoder side, is optional or impossible.

It should be noted that the method described herein may be implemented without one or more of the four steps 430, 440, 450, and 470 shown in FIG. 4.

IV. Equivalents, Extensions, Alternatives and Miscellaneous Provisions

Although the present application describes and shows specific embodiments, the invention is not limited to these specific examples. Modifications and changes can be made to the above described embodiments without departing from the scope of the invention, which is defined only by the appended claims.

In the claims, the word "comprising" does not exclude other elements or steps, and the use of the singular does not exclude many. The mere fact that certain measurements are indicated in mutually different dependent claims does not mean that a combination of these measurements cannot be used if necessary. Any reference positions in the claims should not be considered as limiting the scope of the invention.

The devices and methods described above may be implemented in software, firmware, hardware, or a combination thereof. In a hardware implementation, the division of tasks between the functional blocks mentioned in the description above does not necessarily correspond to the division into physical blocks; on the contrary, one physical component may have multiple functionalities, and one task may be performed in a distributed manner by several physical components together. Certain or all of the components may be implemented as software executable by a digital processor, signal processor, or microprocessor, or may be implemented as hardware or as an ASIC. Such software may be distributed on computer-readable media, which may include computer storage media (or non-volatile media) and communications media (or intermediate media). As is well known to those skilled in the art, the term "computer storage media" includes both volatile and non-volatile, removable and non-removable storage media implemented in any method or technology for storing information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, without limitation, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVD) or other optical disc drives, magnetic cassettes, magnetic tape drives, magnetic disks, or other magnetic storage devices, or any other storage medium that can be used to store the desired information and that can be accessed by a computer. In addition, the person skilled in the art is well aware that communication media typically implements computer readable instructions, data structures, program modules, or other data in the form of a modulated data signal, such as a carrier wave or other transfer mechanism, and includes some means of delivering information.

Claims (39)

1. An audio decoding method (1200), including: receiving (1201) a two-channel downmix signal (L ₁ , L ₂ ) and upmix parameters to reconstruct the M -channel audio signal based on the downmix signal; receiving (1202) a signaling (S) indicative of a selected one of at least two M -channel audio coding formats, the coding formats corresponding to the M -channel audio channels divided into respective first and second groups (601, 602) of one or more channels, wherein, in said coding format, the first channel of the downmix signal corresponds to a first linear combination of one or more channels of the M -channel audio signal, and the second channel of the downmix signal corresponds to a second linear combination of one or more channels of the M -channel audio signal; calculating the first upmix signal as a linear mapping of the downmix signal, the first set of upmix coefficients being applied to the downmix signal; calculating a second upmix signal as a linear display of the downmix signal, the second set of upmix coefficients being applied to the downmix signal; And combining (1213) the first and second upmix signals to obtain a multi-dimensional reconstructed signal corresponding to the M -channel audio signal to be restored, characterized in that The M -channel audio signal has a predetermined channel configuration, and said selected coding format is switched between at least two coding formats. 2. An audio decoding method according to claim 1, further comprising determining the existence of a set of coefficients based on the specified coding format. 3. Method for decoding sound according to any one of paragraphs. 1, 2, characterized in that the set of coefficients is adapted based on the M -channel audio signal. 4. The method of decoding sound according to any one of paragraphs. 1-3, characterized in that it additionally includes: determining (1203) a set of pre-decoration coefficients based on the specified coding format; calculating (1205) the input decorrelation signal (D ₁ , D ₂ , D ₃ ) as a linear mapping of the downmix signal, wherein a set of pre-decoration coefficients are applied to the downmix signal, the pre-decoration coefficients being determined such that the first channel (TBL) of the M -channel the audio signal contributes, via the downmix signal, to the first fixed channel (D3) of the decorrelation input signal in at least two coding formats; generating (1207) a decorrelated signal based on the input decorrelation signal; And determining the second upmix signal as a linear mapping of the decorrelated signal. 5. Method for decoding sound according to any one of paragraphs. 1-4, characterized in that each of the input decorrelation signal and the decorrelated signal contains M -2 channels, and the channel of the decorrelated signal is generated based on no more than one channel of the decorrelation input signal, and the pre-decrelation coefficients are determined in such a way that in each of the coding formats, the decorrelation input signal channel receives a contribution from at most one downmix signal channel. 6. Method for decoding sound according to any one of paragraphs. 1-5, characterized in that the pre-decrelation coefficients are determined in such a way that additionally the second channel (L) of the M -channel audio signal contributes via the downmix signal to the second fixed channel (D1) of the input decorrelation signal in at least two coding formats; and/or wherein the pre-decrelation coefficients are determined such that a channel pair (LS, LB) of the M -channel audio signal contributes via a downmix signal to a third fixed channel (D2) of the decorrelation input signal in at least two coding formats. 7. Method for decoding sound according to any one of paragraphs. 1-6, characterized in that it additionally includes: responsive to detecting a switch of said coding format from the first coding format to the second coding format, performing (1206) a gradual transition from pre-decoration coefficient values associated with the first coding format to pre-decoration coefficient values associated with the second coding format. 8. Method for decoding sound according to any one of paragraphs. 1-7, characterized in that, in response to the detection of the received signaling, indicating the first predefined channel configuration, further includes: receiving a two-channel downmix signal (L ₁ , L ₂ ) and associated upmix parameters (α); And performing parametric reconstruction of the first audio signal based on the first channel (L ₁ ), the downmix signal, and at least some of the upmix parameters. 9. Method for decoding sound according to any one of paragraphs. 1-8, characterized in that, in response to detecting the received signaling indicative of the first predetermined channel configuration, further provides for performing parametric reconstruction of the second audio signal based on the second channel (L ₂ ), the downmix signal, and at least some of the parameters upmix. 10. An audio decoding system (1000), comprising one or more components configured to perform the method of claim 1. 11. The audio decoding system according to claim 10, characterized in that it additionally provides for determining the existence of a set of coefficients based on the specified coding format. 12. The audio decoding system according to any one of paragraphs. 10, 11, characterized in that the set of coefficients is adapted based on the M -channel audio signal. 13. The audio decoding system according to any one of paragraphs. 10-12, characterized in that it additionally provides: determining (1203) a set of pre-decoration coefficients based on the specified coding format; calculating (1205) the input decorrelation signal (D ₁ , D ₂ , D ₃ ) as a linear mapping of the downmix signal, wherein a set of pre-decoration coefficients are applied to the downmix signal, the pre-decoration coefficients being determined such that the first channel (TBL) of the M -channel the audio signal contributes, via the downmix signal, to the first fixed channel (D3) of the decorrelation input signal in at least two coding formats; generating (1207) a decorrelated signal based on the input decorrelation signal; And determining the second upmix signal as a linear mapping of the decorrelated signal. 14. The audio decoding system according to any one of paragraphs. 10-13, characterized in that each of the input decorrelation signal and the decorrelated signal contains M -2 channels, and the channel of the decorrelated signal is generated based on no more than one channel of the decorrelation input signal, and the pre-decrelation coefficients are determined in such a way that in each of the coding formats, the decorrelation input signal channel receives a contribution from at most one downmix signal channel. 15. The audio decoding system according to any one of paragraphs. 10-14, characterized in that the pre-decrelation coefficients are determined such that additionally the second channel (L) of the M -channel audio signal contributes via the downmix signal to the second fixed channel (D1) of the input decorrelation signal in at least two coding formats; and/or wherein the pre-decrelation coefficients are determined such that a channel pair (LS, LB) of the M -channel audio signal contributes via a downmix signal to a third fixed channel (D2) of the decorrelation input signal in at least two coding formats. 16. The audio decoding system according to any one of paragraphs. 10-15, characterized in that it additionally provides: responsive to detecting a switch of said coding format from the first coding format to the second coding format, performing (1206) a gradual transition from pre-decoration coefficient values associated with the first coding format to pre-decoration coefficient values associated with the second coding format. 17. The audio decoding system according to any one of paragraphs. 10-16, characterized in that, in response to the detection of the received signaling indicating the first predetermined channel configuration, further provides: receiving a two-channel downmix signal (L ₁ , L ₂ ) and associated upmix parameters (α); And performing parametric reconstruction of the first audio signal based on the first channel (L ₁ ), the downmix signal, and at least some of the upmix parameters. 18. The audio decoding system according to any one of paragraphs. 10-17, characterized in that, in response to detecting the received signaling indicative of the first predetermined channel configuration, it further provides for performing parametric reconstruction of the second audio signal based on the second channel (L ₂ ), the downmix signal, and at least some of the parameters upmix.

Images