JP4728398B2 - Audio signal decoding method and apparatus - Google Patents

Description

  The present invention relates to audio signal processing, and more particularly, to an audio signal decoding method and apparatus for decoding an audio signal.

  In general, when an encoding apparatus encodes an audio signal and the encoding audio signal is a multi-channel audio signal, the multi-channel audio signal is down-mixed into two channels or one channel to generate a down-mix audio signal, Extract spatial information from multi-channel audio signals. This spatial information is information that can be used to upmix a downmix audio signal to a multi-channel audio signal. Meanwhile, the encoding device downmixes the multi-channel audio signal according to a predetermined tree structure. Here, the determined tree structure may be a structure (s) promised between the audio signal decoding apparatus and the audio signal encoding apparatus. That is, if there is only identification information indicating which type corresponds to a predetermined tree structure, the decoding apparatus can perform the structure of the audio signal after upmixing, for example, the number of channels and the position of each channel. I understand.

  As described above, when the encoding apparatus downmixes the multi-channel audio signal according to the determined tree structure, the spatial information extracted in this process also depends on the structure. Therefore, when the decoding apparatus upmixes a downmix audio signal using spatial information depending on the structure, a multichannel audio signal having the structure is generated.

  That is, when the decoding device uses the spatial information generated by the encoding device as it is, it is upmixed only to the structure promised by the encoding device and the decoding device, so that an output channel audio signal other than the promised structure is generated. There was a problem of not being. For example, it cannot be upmixed to an audio signal with a number of channels that is different (small or large) from the number of channels determined by the promised structure.

  The present invention has been made to solve the above problems, and an object thereof is to provide an audio signal decoding method and apparatus that can decode an audio signal in a structure other than the structure determined by the encoding apparatus. There is.

  Another object of the present invention is to provide an audio signal decoding method and apparatus capable of decoding an audio signal using the deformed spatial information after the spatial information generated by the encoding is deformed. is there.

  According to one aspect of the present invention for achieving the above object, a step of receiving an audio signal and spatial information, a step of identifying a type of modified spatial information, and the modified spatial information using the spatial information. And decoding the audio signal using the deformation space information, and the type of the deformation space information includes subspace information, combination space information, and expanded space information. An audio signal decoding method comprising one or more is provided.

  According to another aspect of the present invention, receiving spatial information, generating combined spatial information using the spatial information, and decoding an audio signal using the combined spatial information. In addition, there is provided an audio signal decoding method, wherein the combined spatial information is generated by combining spatial parameters included in the spatial information.

  According to still another aspect of the present invention, receiving spatial information including one or more spatial parameters and spatial filter information including one or more filter parameters; combining the spatial parameters and the filter parameters; There is provided a method of decoding an audio signal, comprising: generating combination space information having a surround effect; and converting the audio signal into a virtual surround signal using the combination space information.

  According to still another aspect of the present invention, a step of receiving an audio signal, a step of receiving spatial information including tree structure information and a spatial parameter, and adding extended spatial information to the spatial information to obtain modified spatial information. Generating an audio signal using the modified spatial information, wherein the upmixing includes converting the audio signal into a primary upmixing signal based on the spatial information. And a method of decoding an audio signal, comprising: converting the primary upmixing audio signal into a secondary upmixing audio signal based on the extended spatial information.

  Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the terms and words used in this specification and claims should not be construed to be limited to ordinary or dictionary meanings, but the inventor should use terms to describe his invention in the best possible way. Based on the principle that this concept can be appropriately defined, it must be interpreted as a meaning and concept consistent with the technical idea of the present invention. Therefore, the embodiments described in the present specification and the configuration shown in the drawings are only the most preferred embodiments of the present invention, and do not limit the technical idea of the present invention. Obviously, various equivalents and variations are possible to replace.

  The terminology used in the present invention is selected from general terms that are currently widely used as much as possible, but in certain cases, there are terms arbitrarily selected by the applicant. Since the meaning is described in detail in the explanation part of the invention, the present invention must be grasped not by a simple term name but by the meaning of the term.

  The present invention generates deformation space information using spatial information, and then decodes an audio signal using the generated deformation space information. Spatial information is spatial information extracted in the process of being downmixed by a predetermined tree structure, and deformed spatial information is spatial information newly generated using the spatial information.

  Hereinafter, the present invention will be specifically described with reference to FIG. FIG. 1 is a diagram illustrating a configuration of an audio signal encoding apparatus and decoding apparatus according to an embodiment of the present invention. Referring to FIG. 1, an audio signal encoding apparatus (hereinafter referred to as an encoding apparatus) 100 includes a downmix unit 110 and a spatial information extraction unit 120, and an audio signal decoding apparatus (hereinafter referred to as a decoding apparatus). 200 includes an output channel generation unit 210 and a deformation space information generation unit 220.

  The downmix unit 110 of the encoding apparatus 100 generates a downmix audio signal d by downmixing the multichannel audio signal IN_M. The downmix audio signal d may be a signal obtained by downmixing the multichannel audio signal IN_M by the downmix unit 110, but an arbitrary downmix in which the multichannel audio signal IN_M is arbitrarily downmixed by the user. It may be an audio signal.

  The spatial information extraction unit 120 of the encoding apparatus 100 extracts the spatial information s from the multichannel audio signal IN_M. Here, the spatial information is information necessary for upmixing the downmix audio signal d into the multichannel audio signal IN_M. On the other hand, the spatial information may be information extracted in a process in which the multi-channel audio signal IN_M is downmixed by a predetermined tree structure. Here, the defined tree structure is a tree structure (s) promised between the audio signal decoding apparatus and the audio signal encoding apparatus, but the present invention is not limited to this. Meanwhile, the spatial information may include tree structure information, an indicator, a spatial parameter, and the like. Here, the tree structure information refers to information related to the type of tree structure, and the number of multi-channels, the downmix order for each channel, and the like vary depending on the type of tree structure. The indicator is information indicating whether or not extended space information exists. Spatial parameters include a level difference between channels (hereinafter referred to as 'CLD') and a correlation between channels (inter channel coherences) when two or more channels are downmixed into two or less channels. : Hereinafter referred to as “ICC”), channel prediction coefficients (hereinafter referred to as “CPC”), and the like. On the other hand, the spatial information extraction unit 120 can further extract extended spatial information in addition to the spatial information. Here, the extended spatial information is information necessary when the downmix audio signal d is additionally mixed after being upmixed with the spatial parameters, and includes extended channel configuration information and extended spatial parameters. Can do. The extended spatial information described later is not limited to that extracted by the spatial information extraction unit 120.

  On the other hand, the encoding apparatus 100 decodes the downmix audio signal d to generate a downmix audio bitstream, a core codec encoding unit (not shown), and encodes the spatial information s to generate a spatial information bitstream. An information encoding unit (not shown) and a multiplexing unit (not shown) that multiplexes the downmix audio bit stream and the spatial information bit stream to generate a bit stream related to the audio signal can be further provided. However, it is not limited to this.

  The decoding apparatus 200 includes a demultiplexer (not shown) that separates a bit stream related to an audio signal into a downmix audio bitstream and a spatial information bitstream, and a core codec decoding that decodes the downmix audio bitstream. A spatial information decoding unit (not shown) for decoding the spatial information bitstream, but the present invention is not limited to this.

  The modified spatial information generation unit 220 of the decoding apparatus 200 identifies the type of the modified spatial information using the spatial information, and generates modified spatial information s ′ of the type identified based on the spatial information. . Here, the spatial information may be the spatial information s transmitted from the encoding apparatus 100. Modified spatial information refers to spatial information newly generated using spatial information. On the other hand, there are various types of deformation space information (type), and the type of deformation space information may include one or more of a) partial space information, b) combination space information, and c) expanded space information. However, the present invention is not limited to this. The partial space information includes a part of the spatial parameters, the combined space information is generated by combining the spatial parameters, and the expanded spatial information is generated using the spatial information and the extended spatial information. The method by which the deformation space information generation unit 220 generates deformation space information varies depending on the type of deformation space information as described above. The method for generating the deformation space information for each type of deformation space information will be described in detail later.

  On the other hand, the criteria for determining the type of the modified spatial information can be tree structure information in the spatial information, an indicator in the spatial information, output channel information, and the like. The tree structure information and the indicator can be included in the spatial information s from the encoding device. The output channel information is information related to the speaker linked with the decoding apparatus 200, and may include the number of output channels, position information of each output channel, and the like. The output channel information may be already input by the producer or may be input by the user. A method for determining the type of deformation space information using such information will be described in more detail later.

  The output channel generator 210 of the decoding apparatus 200 generates the output channel audio signal OUT_N from the downmix audio signal d using the modified spatial information s ′.

  The spatial filter information 230 of the decoding device 200 is information related to the acoustic path and is provided to the modified spatial information generation unit 220. This spatial filter information can be used when the modified spatial information generation unit 220 generates combined spatial information having a surround effect.

Hereinafter, a method of generating the deformation space information for each type of the deformation space information and decoding the audio signal will be described in the order of (1) partial space information, (2) combination space information, and (3) expanded space information.
(1) Subspace information

  Since the spatial parameter is calculated in the process of downmixing the multi-channel audio signal according to a predetermined tree structure, if the downmix audio signal is decoded using the spatial parameter as it is, the spatial parameter is not changed. Is restored to the original multi-channel audio signal. If it is desired to reduce the number of channels (N) of the output channel audio signal from the number of channels (M) of the multi-channel audio signal, the downmix audio signal can be decoded by applying only a part of the spatial parameters. .

  Such a method can be changed according to the order and method in which the multi-channel audio signal is downmixed by the encoding apparatus, that is, the type of the tree structure, and the type of the tree structure is queried using the tree structure information of the spatial information. can do. Further, in this method, the number of output channels can be changed depending on some, and the number of output channels may be inquired using the output channel information.

Hereinafter, when the number of channels of the output channel audio signal is smaller than the number of channels of the multi-channel audio signal, various tree structures are used for decoding the audio signal by applying subspace information including a part of the spatial parameters. Will be explained.
(1) -1. First example of tree structure (5-2-5 tree structure)

FIG. 2 is a diagram schematically illustrating an example of applying partial space information. Referring to the left side of FIG. 2, a multi-channel audio signal having six channels (left front channel L, left surround channel L _S , center channel C, low frequency channel LFE, right front channel) The order in which the channel (Right Front) R and the right surround channel (Right Surround R _S ) are downmixed to the stereo downmix channels L _o and R _o and the relationship with the spatial parameters are shown.

First, a downmix between left channel L and the left surround channel L _S, downmixing between the center channel C and the low frequency channel LFE, downmixing between the right channel R and the right surround channel R _S is performed. In such a first downmix process, the left total channel L _t , the center total channel C _t , and the right total channel R _t are generated. The spatial parameters calculated in the first downmix process are CLD ₂ (ICC ₂ ), CLD ₁ (including ICC ₁ ), CLD ₀ (including ICC ₀ ), and the like. In the secondary downmix process after the primary downmix process, the left total channel L _t , the center total channel C _t , and the right total channel R _t are _downmixed to generate the left channel _Lo and the right channel _Ro. The spatial parameters calculated in the next downmix process may include CLD _TTT , CPC _TTT , ICC _{TTT, and the} like. In other words, a total of 6 channels of multi-channel audio signals are downmixed according to the above order to generate stereo downmix audio signals L _o and R _o . If spatial parameters (CLD ₂ , CLD ₁ , CLD ₀ , CLD _TTT, etc.) calculated according to such an order are used as they are, they are upmixed in the reverse order of the downmixed order, and the number of channels is six. Audio signals (left front channel L, left surround channel L _S , center channel C, low frequency channel LFE, right front channel R, right surround channel R _S ) are generated.

As shown on the right side of FIG. 2, when the subspace information is CLD _TTT among the spatial parameters (CLD _2, CLD _1, CLD _0, CLD _TTT, etc.), the left total channel L _t , the center total channel C _t , and after upmixing the right total channel R _t, the output channel audio signal as left total channel L _t, by selecting only the right total channel R _t, 2-channel output channel audio signal L _t, can generate R _t, the output When the left total channel L _t , the center total channel C _t , and the right total channel R _t are selected as channel audio signals, three channel output channel audio signals L _t , C _t , R _t can be generated. In addition, if the left total channel L _t , the right total channel R _t , the center channel C, and the low frequency channel LFE are selected as output channel audio signals after upmixing using CLD ₁ , four output channels are selected. Audio signals L _t , R _t , C, and LFE can be generated.
(1) -2. Second example of tree structure (5-1-5 tree structure)

FIG. 3 is a diagram schematically illustrating another example in which the partial space information is applied. Referring to the left side of FIG. 3, a multi-channel audio signal having 6 channels (left front channel L, left surround channel L _S , center channel C, low frequency channel LFE, right front channel R, right surround channel R _S ). The order of downmixing to the mono downmix audio signal M and the relationship with the spatial parameters are shown.

As in the first example of the tree structure, downmixing between the left channel L and the left surround channel L _S , downmixing between the center channel C and the low frequency channel LFE, and down between the right channel R and the right surround channel R _S Mixing is done. In such a first downmix process, the left total channel L _t , the center total channel C _t , and the right total channel R _t are generated, and the spatial parameters calculated in the first downmix process are CLD ₃ (ICC ₃ CLD ₄ (including ICC ₄ ), CLD ₅ (including ICC ₅ ) (here, CLD _x and ICC _x are distinguished from CLD _x in the first example of the tree structure), and the like. In primary downmixing process subsequent secondary downmixing process, the left total channel L _t and the center total channel C _t and is downmixed by the left center channel LC is generated, the center total channel C _t and a right total channel R _t Are mixed down to generate the right center channel RC, and the spatial parameters calculated in the second downmix process are CLD ₂ (including ICC ₂ ), CLD ₁ (including ICC ₁ ), and the like. Thereafter, the left center channel LC and the right center channel RC are downmixed in the third downmix process to generate the mono downmix channel M, and the spatial parameter calculated in the second downmix process is CLD ₀ (ICC ₀ ).

As shown on the right side of FIG. 3, when the subspace information is CLD ₀ among the spatial parameters (CLD ₃ , CLD ₄ , CLD ₅ , CLD ₁ , CLD ₂ , CLD _0, etc.), the left center channel LC and the right center After the channel RC is generated, when the left center channel LC and the right center channel RC are selected as output channel audio signals, two channel output channel audio signals LC and RC can be generated. On the other hand, when the subspace information is CLD ₀ , CLD ₁ , CLD ₂ among the spatial parameters (CLD ₃ , CLD ₄ , CLD ₅ , CLD ₁ , CLD ₂ , CLD _0, etc.), the left total channel L _t , center total After generating the channel C _t and the right total channel R _t , if the left total channel L _t and the right total channel R _t are selected as output channel audio signals, two channel output channel audio signals L _t and R _t can be generated, When the left total channel L _t , the center total channel C _t and the right total channel R _t are selected as the output channel audio signals, three channel output channel audio signals L _t , C _t and R _t can be generated. When the subspace information additionally includes CLD ₄ , after upmixing to the center channel C and the low frequency channel LFE, the left total channel L _t , right total channel R _t , center channel C and When the low frequency channel LFE is selected, four output channel audio signals L _t , R _t , C, and LFE can be generated.

(1) -3. Third example of tree structure (5-1-5 tree structure)
FIG. 4 is a diagram schematically illustrating still another example in which the partial space information is applied. Referring to the left side of FIG. 4, a multi-channel audio signal having 6 channels (left front channel L, left surround channel L _S , center channel C, low frequency channel LFE, right front channel R, right surround channel R _S ) is mono. The order of downmixing to the downmix audio signal M and the relationship with the spatial parameters are shown.

As in the first and second examples of the tree structure, the downmix between the left channel L and the left surround channel L _S , the downmix between the center channel C and the low frequency channel LFE, the right channel R and the right surround channel R _Downmix between _S is performed. In such a first downmix process, the left total channel L _t , the center total channel C _t , and the right total channel R _t are generated, and the spatial parameters are CLD ₁ (including ICC ₁ ), CLD ₂ (including ICC ₂ ). , CLD ₃ (ICC including _3), etc. (CLD _x, ICC _x here is, CLD _x in the first and second examples of a tree structure, as distinguished from ICC _x) is calculated. In the secondary downmix process after the primary downmix process, the left total channel L _t , the center total channel C _t and the right total channel R _t are downmixed to generate the left center channel LC and the right channel R, and the spatial parameters CLD _TTT (including ICC _TTT ) is calculated. Thereafter, in the third downmix process, the left center channel LC and the right channel R are downmixed to generate the mono downmix channel M, and the spatial parameter CLD ₀ (including ICC ₀ ) is calculated.

As shown on the right side of FIG. 4, when the subspace information is CLD ₀ and CLD _TTT among the spatial parameters (CLD ₁ , CLD ₂ , CLD ₃ , CLD _TTT , CLD _0, etc.), the left total channel L _t , center After generating the total channel C _t and the right total channel R _t , if the left total channel L _t and the right total channel R _t are selected as output channel audio signals, two channels of output channel audio signals L _t and R _t can be generated. When the left total channel L _t , the center total channel C _t and the right total channel R _t are selected as the output channel audio signals, three channel output channel audio signals L _t , C _t and R _t can be generated. If the subspace information additionally includes CLD ₂ , after upmixing to the center channel C and the low frequency channel LFE, the left total channel L _t , the right total channel R _t , the center channel C and the output channel audio signal are output. When the low frequency channel LFE is selected, four output channel audio signals L _t , R _t , C, and LFE can be generated.

  As described above, the process of generating the output channel audio signal by applying only a part of the spatial parameters by taking up the three types of tree structures has been described. However, the process is not limited to the application of the subspace information as described above. Furthermore, combination space information or expanded space information may be applied. As described above, the process of applying the deformed space information to the audio signal may be performed sequentially and hierarchically, or may be processed collectively and collectively.

(2) Combination spatial information Since the spatial information is calculated in the process of downmixing the multi-channel audio signal according to a defined tree structure, the spatial parameter of the spatial information is used as it is for the downmix audio signal. Decoding, the original multichannel audio signal before being downmixed is restored. If the number M of channels of the multi-channel audio signal is different from the number N of channels of the output channel audio signal, after combining the spatial information to generate new combined spatial information, the downmix audio signal can be upmixed using this. Specifically, a spatial parameter can be substituted into a conversion formula to generate a combined spatial parameter.

  Such a method can be changed according to the order and method in which the multi-channel audio signal is downmixed in the encoding device. The order and method in which the multichannel audio signal is downmixed can be inquired by using the tree structure information of the spatial information. good. Also, in this method, the number of output channels can vary depending on some, but the number of output channels and the like may be inquired using the output channel information.

  In the following, a specific example of a method for transforming spatial information will be described, and then an example for providing a virtual 3D effect will also be described.

(2) -1. General combination space information A method for generating a combination space parameter by combining spatial parameters of spatial information is for upmixing with a tree structure different from the tree structure in the downmix process. Whatever the structure, it is applicable to all downmix audio signals.

  When the multi-channel audio signal is 5.1 channel and the downmix audio signal is 1 channel (mono channel), the process of generating the 2-channel output channel audio signal will be described with reference to the following two examples.

(2) -1-1. Fourth example of tree structure (5-1-5 ₁ tree structure)
FIG. 5 is a diagram schematically illustrating an example of applying the combination space information. As shown on the left side of FIG. 5, the spatial parameters that can be calculated in the process of downmixing the 5.1 channel multi-channel audio signal are CLD _{0 to} CLD ₄ and ICC _{0 to} ICC ₄ (not shown), respectively. I can say that. For example, among the spatial parameters, the inter-channel level difference between the left channel signal L and the right channel signal R is CLD ₃ , the inter-channel correlation is ICC ₃ , and the inter-channel levels of the left surround channel LS and the right surround channel _RS The difference is CLD ₂ and the inter-channel correlation is ICC ₂ .

On the other hand, referring to the right side of FIG. 5, when the left and right channel signals L _t and R _t are generated by applying the combination space parameters CLD _α and ICC _α to the mono downmix audio signal m, the mono channel audio signal is obtained. Stereo output channel audio signals L _t and R _t can be generated directly from m. The combination space parameters CLD _α and ICC _α here can be calculated by combining the space parameters CLD _{0 to} CLD ₄ and ICC _{0 to} ICC ₄ . First, the process of calculating CLD _α in the combination spatial parameters by combining CLD _{0 to} CLD ₄ among the spatial parameters will be described, and then CLD _{0 to} CLD ₄ and ICC _{0 to} ICC ₄ among the spatial parameters are combined. A process of calculating ICC _α in the combination space parameter will be described.

(2) -1-1-a. CLD induction of _alpha First, the CLD _alpha, because the level difference between the left output signal L _t and a right output signal R _t, and substituting the left output signal L _t and a right output signal R _t in the definition formula of CLD, following It becomes as follows.

Ｐ_LtはＬ_tのパワー（ｐｏｗｅｒ）、Ｐ_RtはＲ_tのパワーを表す。

P _Lt represents the power (power) of L _t , and P _Rt represents the power of R _t .

Ｐ_LtはＬ_tのパワー、Ｐ_RtはＲ_tのパワー、ａは非常に小さい定数を表す。

P _Lt is the power of L _t , P _Rt is the power of R _t , and a is a very small constant.

CLD _α is defined as in Formula 1 or Formula 2 above.

On the other hand, in order to express P _Lt and P _Rt using the spatial parameters CLD _{0 to} CLD ₄ , the left output signal L _t , the right output signal R _t, and the multichannel signals L, L _s , R of the output channel audio signal. , R _s , C, and LFE are necessary, and the relation can be defined as follows.

  Since the relational expression such as Expression 3 can change depending on how the output channel audio signal is defined, it is natural that an expression different from Expression 3 can be defined. For example, in Equation 3, the 1 / √2 factor in C / √2 or LFE / √2 can be ‘0’ or ‘1’.

  A relational expression such as the following Expression 4 can be derived from Expression 3.

CLD _α can be expressed using P _Lt and P _{Rt according} to Equation 1 (or Equation 2), and such P _Lt and P _Rt can be expressed as P _L , P _Ls , P _C , P _LFE according to Equation 4. , P _R , P _Rs can be expressed by using the spatial parameters CLD _{0 to} CLD ₄ , P _L , P _Ls , P _C , P _LFE , P _R , P _Rs can be expressed using spatial parameters CLD _{0 to} CLD _4. Need to ask.

On the other hand, in the case of the tree structure as shown in FIG. 5, the relationship among the multi-channel audio signals L, R, C, LFE, L _s , R _s and the mono downmix channel signal m is as follows.

  From equation 5, the following equation 6 can be derived:

That is, by substituting Equation 6 into Equation 4 and Equation 4 into Equation 1 (or Equation 2), CLD _α that is a combination space parameter is expressed by combining CLD _{0 to} CLD ₄ that are space parameters. can do.

On the other hand, the expansion formula obtained by substituting Formula 6 into P _C / 2 + P _LFE / 2 in Formula 4 is as follows.

ここで、ｃ₁及びｃ₂の定義によれば（式５参照）、(ｃ_1,x)²＋(ｃ_2,x)²＝１なので、(ｃ_1,OTT4)²＋(ｃ_2,OTT4)²＝１である。

Here, according to the definitions of c ₁ and c ₂ (see Equation 5), since (c _{1, x} ) ² + (c _{2, x} ) ² = 1, (c _{1, OTT4} ) ² + (c _{2, OTT4} ) ² = 1.

  Therefore, Equation 7 can be simplified as follows.

In short, by substituting Equations 8 and 6 into Equation 4 and Equation 4 into Equation 1, CLD _α , which is a combination space parameter, is expressed in a manner that combines CLD _{0 to} CLD ₄ that are space parameters. be able to.

(2) -1-1-b. Induction of ICC _alpha First, the ICC _alpha, because it is the correlation between the left output signal L _t and a right output signal R _t, and substituting the left output signal L _t and a right output signal R _t in the defining equation, the following It becomes street.

In Expression 9, P _Lt and P _Rt can be expressed by using CLD _{0 to} CLD _{4 according} to Expression 4, Expression 6, and Expression 8, and P _Lt P _Rt is expanded as in Expression 10 below. be able to.

In Equation 10, P _C / 2 + P _LFE / 2 can be expressed as CLD _{0 to} CLD ₄ according to Equation 6, and P _LR and P _LsRs can be expanded as follows according to the ICC definition.

In equation 11, when √ (P _L P _R ) (or √ (P _Ls P _Rs )) is transferred, the following equation 12 is obtained.

In Expression 12, P _L , P _R , P _Ls , and P _Rs can be expressed as CLD _{0 to} CLD _{4 according} to Expression 6, respectively. Substituting Expression 6 into Expression 12, the following Expression 13 is obtained.

In short, by substituting Equation 6 and Equation 13 into Equation 10 and Equation 10 and Equation 4 into Equation 9, ICC _α that is a combinational space parameter becomes CLD _{0 to} CLD ₃ and ICC ₂ that are space parameters. It can be expressed in ICC ₃ .

(2) -1-2. Fifth example of tree structure (5-1-5 ₂ tree structure)
FIG. 6 is a diagram schematically illustrating another example in which the combination space information is applied. As shown on the left side of FIG. 6, spatial parameters that can be calculated in the process of downmixing a 5.1 channel multi-channel audio signal are CLD _{0 to} CLD ₄ and ICC _{0 to} ICC ₄ (not shown), respectively. I can say that. Among the spatial parameters, the inter-channel level difference between the left channel signal L and the left surround channel signal L _s is CLD ₃ , the inter-channel correlation is ICC ₃ , and the inter-channel level difference between the right channel R and the right surround channel _RS Is CLD ₄ and the inter-channel correlation is ICC ₄ .

On the other hand, referring to the right side of FIG. 6, when the left and right channel signals L _t and R _t are generated by applying the combination space parameters CLD _β and ICC _β to the mono downmix audio signal m, the mono channel audio signal is obtained. Stereo output channel audio signals L _t and R _t can be generated directly from m. The combination space parameters CLD _β and ICC _β here can be calculated using the space parameters CLD _{0 to} CLD ₄ and ICC _{0 to} ICC ₄ . First, a process of calculating CLD _β among the combination spatial parameters using CLD _{0 to} CLD ₄ among the spatial parameters will be described, and subsequently, using CLD _{0 to} CLD ₄ and ICC _{0 to} ICC ₄ among the spatial parameters. The process of calculating ICC _β among the combination space parameters will be described.

(2) -1-2-a. CLD induction of _beta First, the CLD _beta, since the level difference between the left output signal L _t and a right output signal R _t, and substituting the left output signal L _t and a right output signal R _t in the defining equation, the following It becomes street.

Ｐ_LtはＬ_tのパワーで、Ｐ_RtはＲ_tのパワーである。

P _Lt is the power of L _t , and P _Rt is the power of R _t .

Ｐ_LtはＬ_tのパワー、Ｐ_RtはＲ_tのパワー、ａは非常に小さい数である。
ＣＬＤ_βは、上記の式１４または式１５のように定義される。

P _Lt is the power of L _t , P _Rt is the power of R _t , and a is a very small number.
CLD _β is defined as in Equation 14 or Equation 15 above.

On the other hand, in order to express P _Lt and P _Rt using the spatial parameters CLD _{0 to} CLD ₄ , the left output signal L _t , the right output signal R _t, and the multichannel signals L, L _s , R of the output channel audio signal. , R _s , C, and LFE are necessary, and the relational expression can be defined as follows in the same manner as Expression 3.

  The relational expression such as Expression 16 can be changed depending on how the output channel audio signal is defined. Therefore, it is natural that other expressions can be defined. For example, 1 / √2 in the C / √2 or LFE / √2 factor can be 0 or 1.

  A relational expression such as the following Expression 17 can be derived from Expression 16.

Formula 14 (or Formula 15), CLD _beta is representable using P _Lt and P _Rt, P _Lt and P _Rt is, P _L in Equation _{15, P Ls, P C,} P LFE, P R since representable using _{P Rs, P L, P Ls} , P C, P LFE, P R, the P _Rs, it is necessary to obtain the relational expression that can be represented using spatial parameters CLD ₀ to CLD _4.

On the other hand, in the case of the tree structure as shown in FIG. 6, the relationship among the multi-channel audio signals L, R, C, LFE, L _s , R _s and the mono downmix channel signal m is as follows.

  From equation 18, a relational expression such as the following equation 19 can be derived.

That is, by substituting equation 19 into equation 17 and substituting equation 17 into equation 14 (or equation 15), CLD _β which is a combination space parameter is a method in which CLD _{0 to} CLD ₄ which are space parameters are combined. Can be expressed.

On the other hand, the expansion equation obtained by substituting Equation 19 for P _L + P _Ls in Equation 17 is as follows.

Here, according to the definitions of c ₁ and c ₂ (see Equation 5), since (c _{1, x} ) ² + (c _{2, x} ) ² = 1, (c _{1, OTT3} ) ² + (c _{2, OTT3} ) ² = 1.

  Therefore, Equation 20 can be simplified as follows.

On the other hand, the expansion equation obtained by substituting Equation 19 for P _R + P _Rs in Equation 17 is as follows.

Here, according to the definition of c ₁ and c ₂ (see Equation 5), since (c _{1, x} ) ² + (c _{2, x} ) ² = 1, (c _{1, OTT4} ) ² + (c _{2, OTT4} ) ² = 1.

  Therefore, Equation 22 can be simplified as follows.

On the other hand, the expansion equation obtained by substituting Equation 19 into P _C / 2 + P _LFE / 2 in Equation 17 is as follows.

ここで、ｃ₁及びｃ₂の定義によれば（式５参照）、(ｃ_1,x)²+(ｃ_2,x)²＝１なので、(ｃ_1,OTT2)²＋(ｃ_2,OTT2)²＝１である。

Here, according to the definitions of c ₁ and c ₂ (see Equation 5), since (c _{1, x} ) ² + (c _{2, x} ) ² = 1, (c _{1, OTT2} ) ² + (c _{2, OTT2} ) ² = 1.

Thus, Equation 24 can be simplified as follows.

In short, by substituting Equation 21, Equation 23, and Equation 25 into Equation 17, and assigning Equation 17 into Equation 14 (or Equation 15), CLD _β , which is a combined space parameter, becomes CLD ₀ through CLD ₀ , which is a spatial parameter. It can be expressed by combining CLD ₄ .

(2) -1-2-b. Induction of ICC _beta First, the ICC _beta, because it is the correlation between the left output signal L _t and a right output signal R _t, and substituting the left output signal L _t and a right output signal R _t in the defining equation, the following It becomes street.

In Expression 26, P _Lt and P _Rt can be expressed using CLD _{0 to} CLD _{4 according} to Expression 19, and P _Lt P _Rt can be expanded as in Expression 27 below.

In Expression 27, P _C / 2 + P _LFE / 2 can be expressed as CLD _{0 to} CLD _{4 according} to Expression 19, and P _{L_R_} can be expanded as follows according to the ICC definition.

When √ (P _{L_P R_} ) is shifted, the following Expression 29 is obtained.

In Expression 29, P _{L_} and P _{R_} can be expressed as CLD _{0 to} CLD _{4 according} to Expression 21 and Expression 23, respectively. When Expression 21 and Expression 23 are substituted into Expression 29, the following Expression 30 is obtained.

In short, by substituting equation 30 into equation 27 and substituting equation 27 and equation 17 into equation 26, ICC _β , which is a combination space parameter, is a method in which CLD _{0 to} CLD ₄ and ICC ₁ that are space parameters are combined. Can be expressed.

The method for transforming the spatial parameter is one embodiment, and the above formula further considers the correlation between channels (eg, ICC ₀ etc.) in addition to the signal energy in determining P _x or P _xy. It can change into various forms depending on the situation.

(2) -2. Combination spatial information having a surround effect When generating a combination space information by combining spatial information, a virtual surround effect can be produced by considering an acoustic path. The virtual surround effect or the virtual 3D effect is an effect in which there is a speaker of a do surround channel even though there is actually no surround channel speaker, for example, via two stereo speakers. This is to output a 5.1 channel audio signal.

  The acoustic path may be spatial filter information, and the spatial filter information may use a function called HRTF (Head-Related Transfer Function), but the present invention is not limited to this. The spatial filter information can include a filter parameter, and the combined spatial parameter can be generated by substituting the filter parameter and the spatial parameter into a conversion formula. Meanwhile, the generated combination space parameter may include filter coefficients.

  Hereinafter, a case where the multi-channel audio signal is five channels and a three-channel output channel audio signal is generated will be described, and a method of considering an acoustic path in order to generate combination spatial information having a surround effect will be described.

FIG. 7 is a diagram showing the positions of the three-channel speakers and the acoustic paths to the speakers and the listener. Referring to FIG. 7, it can be seen that the positions of the three speakers SPK1, SPK2, and SPK3 are the left front L, the center C, and the right R, respectively, and the virtual surround channel positions are the left surround Ls and the right surround Rs. . The acoustic paths from the positions L, C, R of the three speakers and the positions Ls, Rs of the virtual surround channel to the position 1 of the listener's left ear and the position r of the listener's right ear are displayed. Yes.
G _{x_y} displays an acoustic path from the x position to the y position. For example, _{GL_r} represents an acoustic path from the left front position L to the right ear position r of the listener.

If there are speakers in five positions (ie, speakers in left surround (Ls) and right surround (Rs) as well) and the listener is in the position shown in FIG. 7, it flows into the listener's left ear The signal L ₀ and the signal R ₀ flowing into the right ear of the listener are as follows.

Here, L, C, R, Ls, Rs represents a channel for each position, G _{x_y} represents an acoustic path to y position from the x position, * represents convolution.

However, as mentioned above, if there are speakers only at the three positions L, C and R, the signal L _{0_real} flowing into the listener's left ear and the signal R _{0_real} flowing into the listener's right ear are It will be as follows.

  Since the surround channel signals Ls and Rs are not considered in the signal displayed in Expression 32, a virtual surround effect cannot be produced. In order to produce a virtual surround effect, a signal when the left surround channel signal Ls is output from the original position Ls and reaches the listener's positions l and r, and three positions L, which are not the original positions Ls and Rs. What is necessary is just to make it the same as the signal which outputs via the speaker of C and R and arrives at the listener's position l and r. The same applies to the right surround channel signal Rs.

  First, the left surround channel signal Ls will be described. When the left surround channel signal Ls is output from the speaker at the left surround position Ls which is the original position, the left surround channel signal Ls reaches the listener's left ear l and the listener's right ear r. The signals are as follows.

  When the right surround channel signal Rs is output from the speaker at the right surround position Rs, which is the original position, the signals reaching the listener's left ear l and the listener's right ear r are as follows. .

  If the signals reaching the listener's left ear l and listener's right ear r are the same as the components of Equation 33 and Equation 34, they can be output from any position speaker (for example, from the speaker SPK1 at the left front position). The listener can feel as if there are speakers at the left surround position Ls and the right surround position Rs.

  On the other hand, the components displayed in Expression 33 are signals that reach the left ear l of the listener and the right ear r of the listener when output from the speaker at the left surround position Ls. When the component SP is output from the left front speaker SPK1, the signals reaching the listener's left ear 1 and listener's right ear r are as follows.

In Expression 35, ' _{GL_l} ' (or ' _{GL_r} '), which is a component corresponding to the acoustic path from the left front position L to the listener's left ear l (or right ear r), is added. However, the signal reaching the listener's left ear l and the listener's right ear r must be the component displayed in Equation 33, not the component displayed in Equation 35. For this reason, when reaching the listener by outputting from the speaker at the left front position L, a component “ _{GL_l} ” (or “ _{GL_r} ”) is added. If the output from the L speaker SPK1 is, 'G _{L_l'} (or, 'G _{L_r')} to the acoustic path inverse function of 'G _{L_l} ^-1' (or, 'G _{L_r} ^-1') to be taken into consideration Don't be. In other words, when the component corresponding to Expression 33 is output from the speaker SPK1 at the left front position L, it must be transformed as the following expression.

  When a component corresponding to Expression 34 is output from the speaker SPK1 at the left front position L, it must be transformed as the following expression.

  Therefore, the signal L ′ output from the speaker SPK1 at the left front position L can be summarized as follows.

(Ls*G_{Ls_r}*G_{L_r} ^-1及びRs*G_{Rs_r}*G_{L_l} ^-1成分は省略される。）

(Ls * G _{Ls_r} * G _{L_r} ⁻¹ and Rs * G _{Rs_r} * G _{L_l} ⁻¹ components are omitted.)

Signal displayed in Equation 38 reaches the left ear l position of hand listening is output from the speaker SPK1 the left front position, since the acoustic path 'G _{L_l'} factor is added, in the formula 38 'G _{L_l} ^{- The 1} 'term is canceled, resulting in the factors displayed in Equation 33 and Equation 34 remaining.

  FIG. 8 is a diagram illustrating signals output from each position for the virtual surround effect.

  Referring to FIG. 8, when the signals Ls and Rs output from the surround positions Ls and Rs are included in the signal L ′ output from each speaker position SPK1 in consideration of the acoustic path, Equation 38 is obtained. I understand.

In Equation 38, when G _{Ls — l} * G _L ^{— 1 −1} is _simply displayed as HL s — _L , the result is as follows.

  On the other hand, the signal C ′ output from the speaker SPK2 at the center position C can be summarized as follows.

  On the other hand, the signal R ′ output from the speaker SPK3 at the right front position R can be summarized as follows.

FIG. 9 is a diagram conceptually illustrating a method of generating a three-channel signal using a five-channel signal as in Expression 38, Expression 39, and Expression 40. When the two-channel signals R ′ and L ′ are generated using the five-channel signal or the surround channel signals Ls and Rs are not included in the center channel signal C ′, H _{Ls_C} and H _{Rs_C} are 0.

For convenience of implementation may be used G _{x_y} instead of H _{x_y,} may be used H _{x_y} considering crosstalk (cross-talk) or the like, H _{x_y} becomes various modifications sell.

  The above description is an example of combined spatial information having a surround effect, and it is obvious that the spatial information can be transformed into various forms depending on the application method of the spatial filter information. As described above, the signals output from the speakers through the above-described process (in the above example, the left front channel L ′, the right front channel R ′, and the center channel C ′) have combination space parameters in the combination space information, as described above. And can be generated from a downmix audio signal.

(3) Expanded spatial information Expanded spatial information can be generated by adding expanded spatial information to the spatial information. The audio signal can be upmixed using the expanded spatial information. The upmixing step converts the audio signal into a primary upmixed audio signal based on the spatial information, and performs a primary up based on the extended spatial information. The mixing audio signal is converted into a secondary upmixing audio signal.

  Here, the extended space information may include extended channel configuration information, extended channel mapping information, and extended space parameters. The extended channel configuration information is information about channels that can be configured in addition to the channels that can be configured by the tree structure information of the spatial information, and can include one or more of a divided identifier and an undivided identifier. A specific description thereof will be described later. The extended channel mapping information is position information of each channel constituting the extended channel. The extended space parameter is information necessary for one channel to be upmixed to two or more channels, and can include an inter-channel level difference.

  Such extended spatial information may be i) generated by the encoding device and then included in the spatial information, or ii) generated by the decoding device itself. When the extended spatial information is generated by the encoding device, the presence or absence of the extended spatial information can be determined based on the indicator of the spatial information. When the extended spatial information is generated by the decoding device itself, the extended spatial parameter of the extended spatial information may be calculated using the spatial parameter of the spatial information.

  On the other hand, the process of upmixing the audio signal using the expanded spatial information generated based on the spatial information and the extended spatial information may be performed sequentially and hierarchically, but is processed collectively and collectively. May be. If the expanded spatial information can be calculated as a single matrix based on the spatial information and the extended spatial information, the downmix audio signal can be directly and collectively increased to a multi-channel audio signal by using the matrix. You can mix. At this time, the factor which comprises a matrix should just be defined by the spatial parameter and the extended spatial parameter.

  First, a case where extended space information generated by an encoding device is used will be described, and then a case where extended space information is generated by the decoding device itself will be described.

(3) -1: When using the extended spatial information generated by the encoding device: arbitrary tree configuration
First, the extended spatial information is generated by adding the extended spatial information to the spatial information and is generated by the encoding device, and a case where the decoding device receives the extended spatial information will be described. On the other hand, the extended space information here may be information extracted in the process of downmixing the multichannel audio signal by the encoding apparatus.

  First, as described above, the extended space information includes extended channel configuration information, extended channel mapping information, and extended space parameters. Here, the extended channel configuration information includes one or more divided identifiers and undivided identifiers. Hereinafter, the process of configuring the extension channel based on the arrangement of the divided identifier and the undivided identifier will be specifically described.

  FIG. 10 is a diagram illustrating an example in which an extended channel is configured based on extended channel configuration information. Referring to the lower part of FIG. 10, 0 and 1 are repeatedly arranged in order, where 0 represents an undivided identifier and 1 represents a divided identifier. First, the undivided identifier 0 exists in the first (1), and the channel matched with the first undivided identifier 0 is the left channel L existing at the uppermost end. Therefore, the left channel L matched with the undivided identifier 0 is selected as an output channel without being divided. In the second (2), the division identifier 1 exists, and the channel matched with the second division identifier 0 is the left surround channel Ls next to the left channel L. Therefore, the left surround channel Ls matched with the division identifier 1 is divided into two channels. Since the undivided identifier (0) exists in the third (3) and the fourth (4), the two channels divided from the left surround channel Ls are not divided and selected as output channels as they are. By repeating such a process up to the last order (10), the entire extended channel is configured.

The channel division process is repeated by the number of division identifiers 1, and the process of selecting a channel as an output channel is repeated by the number of undivided identifiers 0. Therefore, the number of channel division units AT ₀ and AT _{1 is the same as} the number of division identifiers 1 (two), and the number of extension channels L, Lfs, Ls, R, Rfs, Rs, C, and LFE are not divided. This is the same as the number of identifiers 0 (eight).

  On the other hand, after configuring the extended channel, the position can be mapped again for each output channel using the extended channel mapping information. In the case of FIG. 10, the left front channel L, the left front side channel Lfs, the left surround channel Ls, the right front channel R, the right front side channel Rfs, the right surround channel Rs, the center channel C, and the low frequency channel LFE are mapped in this order. .

As described above, an extended channel can be configured based on the extended channel configuration information, and a channel dividing unit for dividing one channel into two or more channels is necessary. When the channel dividing unit divides one channel into two or more channels, an extended space parameter can be used. Since this extended space parameter is the same as the number of channel division units, it is the same as the number of division identifiers. Accordingly, the extended space parameters can be extracted by the number of division identifiers. FIG. 11 is a diagram showing the configuration of the extension channel shown in FIG. 10 and the relationship with the extension space parameters. Referring to FIG. 11, there are two channel division units AT ₀ and AT ₁ , and the extended space parameters ATD ₀ and ATD ₁ respectively applied thereto are displayed. When the extension space parameter is an inter-channel level difference, the channel dividing unit can determine the level of each channel divided into two using the extension space parameter. In the process of adding the extended space information and performing upmixing as described above, only a part of the extended space parameters may be applied instead of the whole.

(3) -2 When generating extended space information: interpolation / extrapolation
The extended spatial information can be generated by adding the extended spatial information to the spatial information, and a case where the extended spatial information is generated using the spatial information will be described. Extended spatial information can be generated using spatial parameters in the spatial information. In this case, a method such as interpolation or extrapolation can be used.

(3) -2-Expansion to 1.6.1 Channels A case where a 6.1 channel output channel audio signal is to be generated when the multichannel audio signal is 5.1 channel will be described.

  FIG. 12 is a diagram showing the position of the 5.1 channel multi-channel audio signal and the position of the 6.1 channel output channel audio signal. Referring to FIG. 12A, the channel positions of the 5.1 channel multi-channel audio signal are the left front channel L, right front channel R, center channel C, low frequency channel LFE (not shown), left side, respectively. It can be seen that the surround channel Ls and the right surround channel Rs. If the 5.1 channel multi-channel audio signal is a downmixed audio signal, applying only the spatial parameter to the downmix audio signal will increase the 5.1 channel multichannel audio signal again. To be mixed. However, as shown in (b) of FIG. 12, in order to upmix a 6.1 channel multi-channel audio signal, a rear center RC channel signal must be further generated.

  The channel signal of the rear center RC can be generated using spatial parameters associated with the two rear channels (the left surround channel Ls and the right surround channel Rs). Specifically, among the spatial parameters, the inter-channel level difference (CLD) represents the level difference between the two channels, and the position of the virtual sound source existing between the two channels can be determined by adjusting the level difference between the two channels. Can be changed.

  Hereinafter, the principle of changing the position of the virtual sound source due to the level difference between the two channels will be described.

  FIG. 13 is a diagram showing the relationship between the level difference between two channels and the position of the virtual sound source. In FIG. 13, the level of the left surround channel Ls is a, and the level of the right surround channel Rs is b. Referring to FIG. 13A, when the level a of the left surround channel Ls is higher than the level b of the right surround channel Rs, the virtual sound source position VS is higher than that of the right surround channel Rs. You can see that it is close to the position. When audio signals are output from two channels, the listener feels as if a virtual sound source exists between the two channels. At this time, the position of the virtual sound source is relative to the level of the two channels. Close to high channel position. In the case of FIG. 13B, since the level a of the left surround channel Ls is substantially the same as the level b of the right surround channel Rs, does the virtual sound source position exist between the left surround channel Ls and the right surround channel Rs? The listener will feel like this.

  Using such a principle, the level of the rear center RC can be determined. FIG. 14 is a diagram illustrating the levels of two rear channels and the level of the rear center channel. As shown in FIG. 14, the level c of the rear center channel RC can be calculated by interpolating the difference between the level a of the left surround channel Ls and the level b of the right surround channel Rs. As the interpolation method, not only linear interpolation but also non-linear interpolation method can be applied. An equation for calculating the level c of a new channel (eg, rear center channel RC) existing between two channels (eg, Ls, Rs) by the linear interpolation method is as follows.

[Formula 42]
c = a * k + b * (1-k)

  Here, a and b represent the levels of the two channels.

  k represents a relative position between the a-level channel and the b-level channel and the c-level channel.

  If the c level channel (eg, rear center channel RC) is located at the exact center of the a level channel (eg, Ls) and the b level channel Rs, k is 0.5. When k is 0.5, the equation 42 becomes the following equation.

[Formula 43]
c = (a + b) / 2

  According to Equation 43, if the c-level channel (eg, the rear center channel RC) is located in the middle of the a-level channel (eg, Ls) and the b-level channel Rs, the new channel level c is The average value of the levels a and b of the channels. The above equations 42 and 43 are merely examples, and not only the determination of the c level but also the values of the a level and the b level can be readjusted.

(3) -2-2.7.1 Extension to Channel A case where an output channel audio signal of 7.1 channel is to be generated when the multi-channel audio signal is 5.1 channel will be described.

  FIG. 15 is a diagram showing the position of the 5.1 channel multi-channel audio signal and the position of the 7.1 channel output channel audio signal. Referring to (a) of FIG. 15, the channel positions of the 5.1 channel multi-channel audio signal are the left front channel L, the right front channel R, the center channel C, the low frequency, respectively, as in FIG. 12 (a). It can be seen that the channel LFE (not shown), the left surround channel Ls, and the right surround channel Rs. If such a 5.1 channel multi-channel audio signal is a downmixed audio signal, applying only a spatial parameter to the downmix audio signal will similarly result in a 5.1 channel multichannel audio signal. Upmixed. However, in order to upmix the 7.1-channel multi-channel audio signal as shown in FIG. 15B, the left front side channel Lfs and the right front side channel Rfs must be further generated.

  Since the left front side channel Lfs is located between the left front channel L and the left surround channel Ls, the level of the left front side channel Lfs is interpolated using the level of the left front channel L and the level of the left surround channel Ls. Can be determined. FIG. 16 is a diagram illustrating the levels of the two left channels and the level of the left front side channel Lfs. Referring to FIG. 16, it can be seen that the level c of the left front side channel Lfs is a value interpolated linearly based on the level a of the left front channel L and the level b of the left surround channel Ls.

  On the other hand, the left front side channel Lfs is located between the left front channel L and the left surround channel Ls, but is located outside the left front channel L, the center channel C, and the right front channel R. Therefore, the level of the left front side channel Lfs may be determined by extrapolation using the level of the left front channel L, the level of the center channel C, and the level of the right front channel R. FIG. 17 is a diagram illustrating the levels of the three front channels and the level of the left front side channel. Referring to FIG. 17, the level d of the left front side channel Lfs is a value extrapolated linearly based on the level a of the left front channel L, the level c of the center channel C, and the level b of the right front channel R. It can be seen that it is.

  In the above two cases, the process of generating the output channel audio signal by adding the extended spatial information to the spatial information has been described. As described above, in the process of adding and mixing the extended space information, only a part of the extended space parameters may be applied instead of the whole. Thus, the process of applying the spatial parameter to the audio signal may be performed sequentially or hierarchically, but may be processed collectively or collectively.

  According to one aspect of the present invention, an audio signal having a structure different from a predetermined tree structure can be generated, and therefore, an audio signal having various structures can be generated.

  According to another aspect of the present invention, since an audio signal having a structure different from a predetermined tree structure can be generated, even when the number of multi-channels before being downmixed is larger or smaller than the number of speakers, It is possible to generate the same number of output channels as the number of speakers from the downmix audio signal.

  According to still another aspect of the present invention, when generating fewer output channels than the number of multi-channels, an up-channel audio signal is output from the multi-channel audio signal after up-mixing from the down-mix audio signal to the multi-channel audio signal. Since the multi-channel audio signal is generated directly from the downmix audio signal rather than downmixing, it is possible to remarkably reduce the amount of calculation required for decoding the audio signal.

  According to still another aspect of the present invention, since the acoustic path can be considered in generating the combination space information, a surround effect can be obtained virtually even in a situation where a surround channel cannot be output.

1 is a configuration diagram of an audio signal encoding apparatus and decoding apparatus according to the present invention; FIG. It is a figure which shows roughly an example which applies partial space information. It is a figure which shows roughly the other example to which partial space information is applied. It is a figure which shows schematically the further another example to which partial space information is applied. It is a figure which shows roughly an example which applies combination space information. It is a figure which shows roughly the other example to which combination space information is applied. It is a figure which shows the position of a 3 channel speaker, and the acoustic path from a speaker to a listener. It is a figure which shows the signal output from each position of a speaker for a surround effect. It is a figure which shows notionally the method of producing | generating a 3 channel signal using a 5 channel signal. It is a figure which shows an example in which an extended channel is comprised based on extended channel structure information. It is a figure which shows the structure of the extended channel shown in FIG. 10, and the relationship with an extended space parameter. It is a figure which shows the position of a 5.1 channel multi-channel audio signal, and the position of a 6.1 channel output channel audio signal. It is a figure which shows the relationship between the level difference between two channels, and the position of a virtual sound source. It is a figure which shows the level of two back channels, and the level of a back center channel. It is a figure which shows the position of a 5.1 channel multi-channel audio signal, and the position of a 7.1 channel output channel audio signal. It is a figure which shows the level of two left side channels, and the level of a left front side channel (Lfs). It is a figure which shows the level of three front channels, and the level of a left front side channel (Lfs).

Claims (18)

Receiving spatial information including one or more spatial parameters and spatial filter information including one or more filter parameters;
Combining at least one spatial parameter and at least one filter parameter to generate combined spatial information for a surround effect;
Converting an audio signal into a virtual surround signal using the combined spatial information;
A method for decoding an audio signal, comprising:   The method of claim 1, wherein the generating comprises substituting the at least one spatial parameter and the at least one filter parameter into a transformation formula to generate a combined spatial parameter of the combined spatial information. The audio signal decoding method described.   The method of claim 2, wherein the combination space parameter includes a filter coefficient.   The method of claim 2, wherein the conversion formula is determined by tree structure information of the audio signal.   The method of claim 2, wherein the conversion formula is determined by output channel information.   The method of claim 1, wherein the spatial filter information is an acoustic path. Modified spatial information generation for receiving spatial information including at least one spatial parameter and spatial filter information including at least one filter parameter, and generating combined spatial information for the surround effect by combining the spatial information and the spatial filter information And
Receiving the audio signal, and an output channel generating unit for converting the audio signal to a virtual surround signal using the combined spatial information,
An audio signal decoding apparatus including: 8. The audio signal according to claim 7, wherein the combination space parameter of the combination space information is generated by substituting the at least one space parameter and the at least one filter parameter into a conversion formula. Coding equipment. The apparatus of claim 8, wherein the combination space parameter includes a filter coefficient.   The apparatus of claim 8, wherein the conversion formula is determined by tree structure information of the audio signal.   The apparatus of claim 7, wherein the conversion formula is determined by output channel information.   The apparatus of claim 7, wherein the spatial filter information is an acoustic path. The step of receiving further receives the audio signal which is one of a monaural signal or a stereo signal;
The audio signal decoding method according to claim 1, wherein the virtual surround signal is a stereo signal with a 3D effect. The audio signal decoding method according to claim 1, wherein the combination space information includes a combination space parameter for converting the audio signal into the virtual surround signal. The audio signal decoding method according to claim 1, wherein the spatial information includes an inter-channel level difference, an inter-channel correlation, and a channel prediction coefficient in a process of downmixing at least two channels into the audio signal. The audio signal decoding device according to claim 7, wherein the audio signal is one of a monaural signal and a stereo signal, and the virtual surround signal is a stereo signal with a 3D effect. The audio signal decoding apparatus according to claim 7, wherein the combination space information includes a combination space parameter for converting the audio signal into the virtual surround signal. The audio signal decoding apparatus according to claim 7, wherein the spatial information includes an inter-channel level difference, an inter-channel correlation, and a channel prediction coefficient in a process of downmixing at least two channels into the audio signal.

Images