CN101197134A - Method and apparatus for eliminating influence of encoding mode switch-over, decoding method and device - Google Patents
Method and apparatus for eliminating influence of encoding mode switch-over, decoding method and device Download PDFInfo
- Publication number
- CN101197134A CN101197134A CNA2007100060089A CN200710006008A CN101197134A CN 101197134 A CN101197134 A CN 101197134A CN A2007100060089 A CNA2007100060089 A CN A2007100060089A CN 200710006008 A CN200710006008 A CN 200710006008A CN 101197134 A CN101197134 A CN 101197134A
- Authority
- CN
- China
- Prior art keywords
- current frame
- decoding
- zero state
- state response
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 73
- 230000004044 response Effects 0.000 claims abstract description 105
- 238000012545 processing Methods 0.000 claims description 21
- 230000000694 effects Effects 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 6
- 230000015572 biosynthetic process Effects 0.000 claims description 4
- 238000003786 synthesis reaction Methods 0.000 claims description 4
- 238000001914 filtration Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 20
- 238000010586 diagram Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 6
- 230000003044 adaptive effect Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
Images
Landscapes
- Compression, Expansion, Code Conversion, And Decoders (AREA)
Abstract
The invention discloses a method for eliminating coding mode switch influence, which comprises the following steps: receiving the current frame and determining the difference of the coding mode adopted in a foregoing frame and a current frame; obtaining a system function of the current frame and computing the zero state response of the system function; and taking the zero state response value as the value of the current frame and coding the value of the current frame. The invention further discloses a device for eliminating coding mode switch influence, which comprises the following parts: a coding unit, a coding module determining unit, a zero state response computing unit and a replacing unit. The invention further discloses a decoding method and a device which is applicable to the coding method for eliminating coding mode switch influence. The method and the device disclosed according to the invention can reduce the complexity of algorithm and improve the coding performance.
Description
Technical Field
The present invention relates to encoding and decoding technologies, and in particular, to a method and an apparatus for eliminating the effect of encoding mode switching, and a decoding method and an apparatus.
Background
With the development of communication technology and the expansion of multimedia services, the requirements on coding technology are continuously increasing. Enhanced adaptive multi-rate wideband (AMR-WB +) coding techniques are currently applied to digital audio coding.
FIG. 1(a) is a schematic block diagram of an enhanced adaptive multi-speed wideband (AMR-WB +) encoder. As shown in fig. 1(a), in the AMR-WB + encoding process, after signal preprocessing is performed on an original input signal, the original input signal is divided into a high-frequency signal and a low-frequency signal by using a filter bank, then the high-frequency signal and the low-frequency signal are encoded by using different encoding modes, the high-frequency signal is encoded by using a high-frequency encoding mode, and the low-frequency signal is encoded by using a low-frequency encoding mode.
In AMR-WB +, the low frequency coding modes include two, Algebraic Code Excited Linear Prediction (ACELP) mode and transform coded excitation (TCX) mode. ACELP is a coding scheme suitable for speech signals and TCX is a coding scheme suitable for music signals. In AMR-WB +, when coding a low frequency signal, switching between two coding modes, ACELP and TCX, is performed by the following method in order to reduce the effect of the previous coding mode on the current coding mode. When the TCX coding mode is applied to the current frame, if the ACELP coding mode is applied to the previous frame, the current frame is processed through a weighting filter to obtain a weighting signal, a Zero Input Response (ZIR) of the first 128 points, which are the first 2 sub-frames of the current frame, is calculated, the zero input response is windowed, and then the windowed zero input response is removed from the weighting signal. Then, the weighted signal from which the windowed zero-input response is removed is subjected to adaptive windowing, transform domain coding, and the like.
In the AMR-WB + encoding process, the zero input response of the first 128 points of the system function H (z) is calculated by using the output state of the filter of the previous frame. Let z (n) be the value of the zero input response, zw(n) is a value obtained by performing ZIR calculation and windowing on h (z), and includes:
zw(n)＝z(n)*w(n) n＝0～2*N-1
where w (N) is a time domain window function, and N is 64, having the form:
w(n)＝1 n＝0～N-1
w(n)＝(2*N-n)/N n＝N～2*N-1
corresponding to the AMR-WB + encoder above, FIG. 1(b) shows a schematic block diagram of an AMR-WB + decoder. In AMR-WB +, the low frequency decoding modes also include two, ACELP mode and TCX mode, corresponding to the low frequency coding mode. As shown in fig. 1(b), in the decoding process of AMR-WB +, when decoding a low-frequency decoded code stream in the decoded code stream, an ACELP mode is used for decoding a code stream encoded by ACELP at an encoding end, and a TCX mode is used for decoding a code stream encoded by TCX at an encoding end. In the low-frequency decoding process, when a TCX decoding mode is adopted for a current frame, if an ACELP decoding mode is adopted for a previous frame, zero input response of a system is calculated in a current code stream, windowing is carried out on the zero input response, and the value obtained after decoding is added with the zero input response after windowing to be used as the value of the current frame.
As described above, in AMR-WB +, when the coding mode is switched from ACELP to TCX, in the TCX coding process, in order to eliminate the influence of the previous frame on the current frame, a zero input response is calculated and windowed, and then the windowed zero input response is subtracted by the weight signal, so that the calculation complexity is high. Corresponding to the encoding process, in AMR-WB +, when the decoding mode is switched from ACELP to TCX, in the TCX decoding process, a zero-input response is calculated, and after windowing the zero-input response, the windowed zero-input response is added to the decoded signal, so that the computational complexity is high.
Disclosure of Invention
The embodiment of the invention provides a method for eliminating the influence of coding mode switching, which reduces the computational complexity.
The embodiment of the invention provides a device for eliminating the influence of coding mode switching, which reduces the computational complexity.
The embodiment of the invention provides a decoding method, which reduces the computational complexity.
The embodiment of the invention provides a decoding device, which reduces the calculation complexity.
The embodiment of the invention provides a method for eliminating the influence of coding mode switching, which comprises the following steps: receiving a current frame and determining that the coding modes adopted by a previous frame and the current frame are different; obtaining a system function of the current frame, and calculating zero state response of the system function; and taking the zero state response value as the value of the current frame, and coding the value of the current frame.
The embodiment of the invention provides a device for eliminating the influence of coding mode switching, which comprises: the device comprises an encoding unit, an encoding mode determining unit, a zero state response calculating unit and a replacing unit; the encoding unit is used for encoding a current frame; the coding mode determining unit is used for determining whether the coding modes adopted by the previous frame and the current frame are the same or not and sending different results to the zero state response calculating unit; the zero state response calculating unit is used for calculating a zero state response in the current frame after receiving the result that the coding modes adopted by the previous frame and the current frame are different, and sending the calculated zero state response value to the replacing unit; the replacing unit is used for receiving the input signal of the current frame, replacing the input signal of the current frame by the zero state response value received from the zero state response calculating unit, and sending the replaced input signal of the current frame to the encoding unit.
The embodiment of the invention provides a decoding method suitable for the coding method for eliminating the coding mode switching influence, which comprises the steps of receiving a decoding code stream of a current frame, and decoding the current frame according to the decoding mode of the current frame; determining that the decoding modes adopted by the previous frame and the current frame are different; the value obtained by decoding the current frame is used as the value of the current frame.
An embodiment of the present invention provides a decoding apparatus, including: the decoding unit, decode the mode determination unit, the processing unit; the decoding unit is used for receiving the decoding code stream of the current frame and decoding the current frame according to the decoding mode of the current frame; the decoding mode determining unit is used for determining whether the decoding modes adopted by the previous frame and the current frame are the same or not and sending different results to the processing unit; and the processing unit is used for receiving the result that the decoding modes adopted by the previous frame and the current frame are different, and then decoding the current frame to obtain a value as the value of the current frame.
According to the method and the device for eliminating the influence of the coding mode switching, provided by the invention, in the coding process of the current frame, if the previous frame adopts a coding mode different from the coding mode of the current frame, in order to eliminate the influence of the previous frame on the current frame, a value of zero state response is calculated as the value of the current frame, so that the algorithm complexity is reduced, and the coding performance can be improved.
According to the decoding method and the decoding device provided by the invention, the coding method for eliminating the influence of coding mode switching can be adapted, and the computational complexity is reduced.
Drawings
FIG. 1(a) is a schematic block diagram of a prior art AMR-WB + encoder;
FIG. 1(b) is a schematic block diagram of a prior art AMR-WB + encoder;
FIG. 2 is a flowchart of a method for eliminating the effect of switching between encoding modes in embodiment 1 of the present invention;
FIG. 3 is a flowchart of a method for eliminating the influence of ACELP coding mode during TCX coding according to embodiment 2 of the present invention;
FIG. 4 is a flowchart of a method for eliminating the effect of ACELP coding mode during TCX coding according to embodiment 3 of the present invention;
FIG. 5 is a schematic diagram of an apparatus for eliminating the effect of coding mode switching in embodiment 4 of the present invention;
FIG. 6 is a flowchart of decoding in embodiment 5 of the present invention;
FIG. 7 is a flow chart of AMR-WB + decoding in embodiment 6 of the present invention;
fig. 8 is a block diagram showing a decoding apparatus in embodiment 7 of the present invention.
Best mode for carrying out the invention
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments.
The embodiment of the invention provides a coding mode switching method, which has the main idea that: in the encoding process of the current frame, if the encoding modes adopted by the previous frame and the current frame are different, in order to eliminate the influence of the previous frame on the current frame, the value of the zero state response of the current frame is calculated as the value of the current frame.
Example 1
This embodiment 1 provides a method for eliminating the influence of the former coding mode on the latter coding mode when there is a switch between the two coding modes during the coding process. In this embodiment 1, the two coding modes are referred to as a first coding mode and a second coding mode, respectively. Fig. 2 is a flowchart of a method for eliminating the effect of switching between encoding modes in the encoding process in embodiment 1, which includes the following steps:
step 201: when the current frame is coded by adopting the second coding mode, judging whether the previous frame is coded by adopting the first coding mode, if so, executing the step 202; otherwise, if the previous frame is encoded in the second encoding mode, step 203 is performed.
The determination in this step is actually that if the coding modes adopted by the current frame and the previous frame are different, step 202 is executed, and if the coding modes adopted by the current frame and the previous frame are the same, step 203 is directly executed.
Step 202: and obtaining a system function of the current frame, calculating a zero state response of the system function, and taking a zero state response value as a value of the current frame.
Wherein the system function is calculated before the current frame. The zero state response of the system function is: and setting the system output of the previous frame to be zero in the system function, and obtaining a response value of the input signal of the current frame through the system function. The specific calculation steps are as follows: and calculating impulse response of the system function, and convolving the input signal of the current frame by using the impulse response to obtain zero state response of the system function.
Here, it is possible to calculate only the zero-state responses of a limited number of sequences of the current frame and replace the values of the limited number of sequences of the current frame with the obtained zero-state response values, without any change in the values of the other sequences in the current frame. Through the step 202, when the current frame adopts the first coding mode, and the current frame adopts the second coding mode, the previous state of the current frame is set to zero when the zero state response is calculated, and then the corresponding value is replaced by the zero state response value, so that the influence of the previous frame on the current frame can be eliminated.
In order to smoothly switch coding, the zero state response value may be windowed, and then the windowed zero state response value is used as the value of the current frame. Wherein, the window function can adopt a window function with monotone increment.
Step 203: and adopting a second coding mode to code the current frame.
In embodiment 1, only the zero-state response needs to be calculated when the influence of the first coding mode is eliminated, so that the calculation complexity is low and the implementation is simple. When the previous frame is coded by the first coding mode and the current frame is coded by the second coding mode, the influence of the previous frame can be eliminated from the current frame, thereby improving the coding performance of the second coding mode.
The method described in this embodiment 1 can be applied to the switching process of any two encoding modes. For example, the first coding mode may be an ACELP mode, and the second coding mode may be a transform domain coding mode, and may specifically be a TCX mode. Next, a method of switching the coding mode will be described in detail with reference to a specific coding example.
Example 2
This embodiment 2 presents a method for eliminating the effect of ACELP coding mode during TCX coding when there is a switch between ACELP and TCX. As shown in fig. 1, after the original input signal is subjected to signal preprocessing, the original input signal is divided into a high-frequency signal and a low-frequency signal by using a filter bank; when the low-frequency signal is coded, an ACELP coding mode is adopted for coding the voice signal, and a TCX coding mode is adopted for coding the music signal. In the process of encoding the low-frequency signal by using the TCX encoding mode, in order to eliminate the influence of the previous frame using the ACELP encoding mode on the current frame, the TCX encoding is performed on the current frame by the following steps, as shown in fig. 3, including the following steps:
step 301: the input signal of the current frame passes through a weighting filter to obtain a weighting signal.
Step 302: judging whether the previous frame adopts the ACELP coding mode for coding, if so, executing the step 303; otherwise, step 305 is performed.
Step 303: at the current frame, a finite number of sequences of zero state responses are computed.
Here, the zero state response is a zero state response of a system function of the current frame. The zero state responses of the finite sequence are: and setting the output state of the previous frame as zero in the system function, and obtaining a response value after the input signals of a limited sequence in the current frame pass through the system function. The specific calculation steps are as follows: and calculating impulse response of a system function, and convolving the input signals of the finite sequences in the current frame by using the impulse response to obtain zero state response of the finite sequences.
Here, the limited number of sequences is a limited number of consecutive sequences before the current frame, and may be several consecutive subframes before, for example, 2 consecutive subframes before, or several consecutive points before. The sequence length for calculating the zero-state response can be set according to practical situations, and the zero-state response of the first 2 subframes, namely the zero-state response of the first 128 points, can be calculated.
The system function in this step is:
wherein,
is a quantized linear filter obtained by linear prediction analysis,
is a synthesis filter that is a filter of the synthesis,is a perceptual weighting filter that is applied to the signal,
is a weight filter.
Step 304: in the current frame, the corresponding finite number of sequences in the weight signal are replaced by the values of the zero state responses of the obtained finite number of sequences.
For example, the current frame includes 4 subframes, and assuming that only the zero-state responses of the first 2 subframes are calculated in step 303, the values of the weight signals in the first 2 subframes are updated to the values of the zero-state responses of the first 2 subframes in step 304, and the weight signals in the last 2 subframes are not updated. In this way, since the partial value of the weight signal is updated with the value of the zero-state response, the influence of the previous frame on the current frame can be eliminated, and the performance loss caused by the previous frame encoding mode on the encoding of the current frame can be reduced.
Step 305: the TCX encoding may be performed on the weight signal, and specifically, adaptive windowing, transform domain encoding, and the like may be performed.
In this step 305, if the previous frame is encoded in ACELP coding mode, the weighting signal of the corresponding value is replaced by a zero-state response value in step 304; if the previous frame was not encoded in the ACELP coding mode, the weighted signal obtained in step 301 by the weighting filter is directly encoded.
When TCX encoding is performed on a low-frequency signal, on the premise that the influence of the previous frame on the current frame is small, in other words, on the premise that the encoding performance allows, the zero-state response may not be calculated, and the weight signal obtained by the weight filter may be directly encoded.
Example 3
This embodiment 3 provides a windowing process that adds a response to a zero state to that of embodiment 2. Fig. 4 is a flowchart of a method for eliminating the effect of a previous frame using the ACELP coding mode on a current frame in the process of coding using the TCX coding mode in this embodiment 3.
Step 401: the input signal of the current frame passes through a weighting filter to obtain a weighting signal.
Step 402: judging whether the previous frame adopts the ACELP coding mode for coding, if so, executing the step 403; otherwise, step 406 is performed.
Step 403: at the current frame, a finite number of sequences of zero state responses are computed.
Step 404: windowing the zero state response values of the finite number of sequences.
Let the number of finite sequences be 2 x N, z (N) be the value of the zero state response, zw(n) is the value of z (n) after windowing, then:
zw(n)＝z(n)*w(n) n＝0～2*N-1
where w (n) is a time domain window function. Here, the windowing of the zero state response value may be performed by using different window functions, for example, the window function w (n) in the time domain may be:
w(n)＝n/N n＝0～N-1
w(n)＝1 n＝N～2*N-1
by windowing, the energy distribution of the sequence can be adjusted, thus enabling smooth switching of the coding and enhancing the coding performance.
Step 405: in the current frame, the corresponding finite sequence in the weight signal is replaced by the obtained zero state response value of the windowed finite sequence.
Step 406: the TCX encoding may be performed on the weight signal, and specifically, adaptive windowing, transform domain encoding, and the like may be performed.
Example 4
Embodiment 4 provides an apparatus for eliminating the influence of coding mode switching during coding. Fig. 5 is a block diagram of an apparatus for removing the effect of coding mode switching during the encoding process, and as shown in fig. 5, the apparatus for removing the effect of coding mode switching during the encoding process includes: the device comprises an encoding unit, an encoding mode determining unit, a zero state response calculating unit and a replacing unit. The encoding unit encodes a current frame; the coding mode determining unit determines whether the coding modes adopted by the previous frame and the current frame are the same or not, and sends different results to the zero state response calculating unit; the zero state response calculating unit calculates the zero state response in the current frame after receiving the result that the coding modes adopted by the previous frame and the current frame are different, and sends the calculated zero state response value to the replacing unit; the replacing unit is used for receiving the input signal of the current frame, replacing the input signal of the current frame by the zero state response value received from the zero state response calculating unit, and sending the replaced input signal of the current frame to the encoding unit.
The apparatus may further comprise a windowing unit. And the windowing unit receives the zero state response value output by the zero state response calculating unit, windows the zero state response value and outputs the windowed zero state response value to the replacing unit. Thus, the replacement unit replaces the input signal of the current frame with the windowed zero state response value.
The apparatus may further comprise a pre-processing unit. The preprocessing unit receives an input signal of a current frame, performs weight filtering on the input signal to obtain a weight signal, and transmits the weight signal to the replacing unit; the replacing unit replaces the weight signal with a zero-state response value and sends the replaced weight signal to the encoding unit.
The apparatus shown in this embodiment 4 is actually an encoding apparatus for encoding with switching between two encoding modes. For example, the apparatus can be used for a TCX encoder in a low frequency coding scheme in an AMR-WB + encoder as shown in FIG. 1.
The above embodiment gives an encoding method capable of eliminating the influence of encoding mode switching, and a decoding method suitable for the above encoding method is given below.
Example 5
This embodiment 5 gives a decoding method suitable for the encoding method in embodiment 1. In this embodiment 5, two decoding modes corresponding to two encoding modes are referred to as a first decoding mode and a second decoding mode, respectively. Fig. 6 is a decoding flowchart in this embodiment 5, which includes the following steps:
step 601: and the decoding end receives the decoding code stream, determines a decoding mode adopted by the current frame, and decodes by adopting the determined decoding mode.
Here, the decoding mode may be determined according to a flag carried in the decoded code stream. If the current frame is coded by adopting the first coding mode, the current frame needs to be decoded by adopting the first decoding mode, and if the current frame is coded by adopting the second coding mode, the current frame needs to be decoded by adopting the second decoding mode.
Step 602: when the current frame is decoded by adopting the second decoding mode, judging whether the previous frame is decoded by adopting the first decoding mode, if so, executing the step 603; otherwise, if the previous frame was decoded in the second decoding mode, step 604 is performed.
Step 603: and taking the value obtained by decoding in the second decoding mode as the value of the current frame, and ending the process.
Step 604: and performing overlap-and-add (overlap-and-add) processing on the value obtained by decoding in the second decoding mode to obtain the value of the current frame, and ending the flow.
Here, overlap-and-add is an existing processing method, and is processing of an overlap region between two frames when decoding modes of the two frames before and after are the same.
In this embodiment 5, corresponding to the encoding end, the first decoding mode may be an ACELP mode, and the second decoding mode may be a transform domain decoding mode, and specifically may be a TCX mode.
Example 6
This embodiment 6 provides a decoding method suitable for the encoding methods of embodiments 2 and 3. When there is a switch between ACELP and TCX coding modes during the coding process, the corresponding decoding flow is shown in fig. 7, and includes the following steps:
step 701: and the decoding end receives the decoding code stream, determines a decoding mode adopted by the current frame, and decodes by adopting the determined decoding mode.
Here, the decoding mode may be determined according to a flag carried in the decoded code stream. If the current frame adopts ACELP coding, the ACELP mode needs to be adopted for decoding, and if the current frame adopts TCX mode coding, the TCX mode needs to be adopted for decoding.
Step 702: when the current frame is decoded by adopting the TCX mode, judging whether the previous frame is decoded by adopting the ACELP mode, if so, executing the step 703, otherwise, executing the step 704.
Step 703: the value decoded by the TCX mode is used as the value of the current frame, and the process goes to step 705.
Here, in the case where the current frame adopts the TCX mode and the previous frame adopts the ACELP mode, there is no overlap area between the two frames, and thus, no other processing is required, thereby reducing the computational complexity of decoding.
Step 704: performing overlap-and-add processing on the value obtained by decoding in the TCX mode to obtain the value of the current frame, and going to step 705.
Step 705: and performing subsequent decoding processing, such as filtering, on the obtained value of the current frame to obtain a decoded and restored signal.
Example 7
Fig. 8 is a block diagram showing a decoding apparatus according to embodiment 7, which is applied to the encoding apparatus according to embodiment 4. As shown in fig. 8, the decoding apparatus includes: the device comprises a decoding unit, a decoding mode determining unit, a processing unit and an overlapping processing unit. The decoding unit receives the decoding code stream of the current frame and decodes the current frame according to the decoding mode of the current frame; the decoding mode determining unit determines whether the decoding modes adopted by the previous frame and the current frame are the same or not, sends different results to the processing unit and sends the same result to the overlapping processing unit; the processing unit receives the result that the decoding modes adopted by the previous frame and the current frame are different, and then decodes the current frame to obtain a value as the value of the current frame; and after receiving the result that the decoding modes adopted by the previous frame and the current frame are the same, the overlapping processing unit performs overlapping processing on the value obtained by decoding the current frame to obtain the value of the current frame.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (16)
1. A method for canceling the effect of coding mode switching, the method comprising:
receiving a current frame and determining that the coding modes adopted by a previous frame and the current frame are different;
acquiring a system function of a current frame, and calculating a zero state response of the system function;
and taking the zero state response value as the value of the current frame, and coding the value of the current frame.
2. The method of claim 1,
after the calculating the zero state response, the method further comprises: windowing the zero state response value;
taking the zero state response value as the value of the current frame: and taking the zero state response value after windowing as the value of the current frame.
3. The method according to claim 1 or 2,
the system function is: the product of the synthesis filter and the weight filter;
the synthesis filter is obtained by linear prediction analysis.
4. The method according to claim 1 or 2,
the zero state response of the system function is calculated as: and setting the system output of the previous frame to be zero in the system function, and calculating a response value of the input signal of the current frame through the system function.
5. The method according to claim 1 or 2,
the calculated zero state response is: zero state responses are computed for a limited number of the current frame sequence.
6. The method of claim 5, wherein the zero state response value as the current frame value is: and taking the zero state response values of the finite sequences as the values of the corresponding finite sequences in the current frame.
7. The method of claim 5, wherein a limited number of the current frame sequences are: the current frame is preceded by a finite number of consecutive sequences.
8. The method of claim 6,
before calculating the zero state response of the system function, the method further comprises: the input signal of the current frame passes through a weight filter to obtain a weight signal;
the taking the zero state response values of the finite number of sequences as the values of the corresponding finite number of sequences in the current frame is: and updating the weight signals of a finite number of sequences in the current frame into zero state response values of the finite number of sequences.
9. The method according to claim 1, wherein the coding mode adopted by the current frame is transform domain coding mode, and the coding mode adopted by the previous frame is algebraic code-excited linear prediction (ACELP) mode.
10. An apparatus for canceling the effect of coding mode switching, the apparatus comprising: the device comprises an encoding unit, an encoding mode determining unit, a zero state response calculating unit and a replacing unit;
the coding unit is used for coding the current frame;
the coding mode determining unit is used for determining whether the coding modes adopted by the previous frame and the current frame are the same or not and sending different results to the zero state response calculating unit;
the zero state response calculating unit is used for calculating a zero state response in the current frame after receiving the result that the coding modes adopted by the previous frame and the current frame are different, and sending the calculated zero state response value to the replacing unit;
the replacing unit is used for receiving the input signal of the current frame, replacing the input signal of the current frame by the zero state response value received from the zero state response calculating unit, and sending the replaced input signal of the current frame to the encoding unit.
11. The apparatus of claim 10, further comprising: a windowing unit;
the windowing unit is used for receiving the zero state response value output by the zero state response calculating unit, windowing the zero state response value and outputting the windowed zero state response value to the replacing unit.
12. The apparatus of claim 10 or 11, further comprising: a pre-processing unit;
the preprocessing unit is used for receiving an input signal of a current frame, performing weight filtering on the input signal to obtain a weight signal and transmitting the weight signal to the replacing unit;
the replacing unit is used for replacing the weight signal with a zero state response value and sending the replaced weight signal to the encoding unit.
13. A decoding method adapted to the encoding method of claim 1, the method comprising:
receiving a decoding code stream of a current frame, and decoding the current frame according to a decoding mode of the current frame;
determining that the decoding modes adopted by the previous frame and the current frame are different;
the value obtained by decoding the current frame is used as the value of the current frame.
14. The method of claim 13,
the decoding mode adopted by the current frame is a transform domain decoding mode, and the encoding mode adopted by the previous frame is an ACELP mode.
15. A decoding apparatus, characterized in that the apparatus comprises: the decoding unit, decode the mode determination unit, the processing unit;
the decoding unit is used for receiving the decoding code stream of the current frame and decoding the current frame according to the decoding mode of the current frame;
the decoding mode determining unit is used for determining whether the decoding modes adopted by the previous frame and the current frame are the same or not and sending different results to the processing unit;
and the processing unit is used for receiving the result that the decoding modes adopted by the previous frame and the current frame are different, and then decoding the current frame to obtain a value as the value of the current frame.
16. The apparatus of claim 15, further comprising: an overlap processing unit;
the decoding mode determining unit is used for sending the same decoding mode result to the overlapping processing unit;
and the overlapping processing unit is used for receiving the result that the decoding modes adopted by the previous frame and the current frame are the same, and then overlapping the value obtained by decoding the current frame to obtain the value of the current frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2007100060089A CN101197134A (en) | 2006-12-05 | 2007-01-23 | Method and apparatus for eliminating influence of encoding mode switch-over, decoding method and device |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200610164454.8 | 2006-12-05 | ||
| CN200610164454 | 2006-12-05 | ||
| CNA2007100060089A CN101197134A (en) | 2006-12-05 | 2007-01-23 | Method and apparatus for eliminating influence of encoding mode switch-over, decoding method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101197134A true CN101197134A (en) | 2008-06-11 |
Family
ID=39547492
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2007100060089A Pending CN101197134A (en) | 2006-12-05 | 2007-01-23 | Method and apparatus for eliminating influence of encoding mode switch-over, decoding method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101197134A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105229734A (en) * | 2013-05-31 | 2016-01-06 | 索尼公司 | Code device and method, decoding device and method and program |
| CN106575507A (en) * | 2014-07-28 | 2017-04-19 | 弗劳恩霍夫应用研究促进协会 | Method and apparatus for processing an audio signal, audio decoder, and audio encoder |
| CN106663442A (en) * | 2014-07-28 | 2017-05-10 | 弗劳恩霍夫应用研究促进协会 | Audio decoder, method and computer program using a zero-input-response to obtain a smooth transition |
| WO2019227931A1 (en) * | 2018-05-31 | 2019-12-05 | 华为技术有限公司 | Method and apparatus for calculating down-mixed signal |
| CN110556116A (en) * | 2018-05-31 | 2019-12-10 | 华为技术有限公司 | Method and apparatus for calculating downmix signal and residual signal |
-
2007
- 2007-01-23 CN CNA2007100060089A patent/CN101197134A/en active Pending
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105229734A (en) * | 2013-05-31 | 2016-01-06 | 索尼公司 | Code device and method, decoding device and method and program |
| CN105229734B (en) * | 2013-05-31 | 2019-08-20 | 索尼公司 | Code device and method, decoding apparatus and method and computer-readable medium |
| CN106575507A (en) * | 2014-07-28 | 2017-04-19 | 弗劳恩霍夫应用研究促进协会 | Method and apparatus for processing an audio signal, audio decoder, and audio encoder |
| CN106663442A (en) * | 2014-07-28 | 2017-05-10 | 弗劳恩霍夫应用研究促进协会 | Audio decoder, method and computer program using a zero-input-response to obtain a smooth transition |
| CN106663442B (en) * | 2014-07-28 | 2021-04-02 | 弗劳恩霍夫应用研究促进协会 | Audio decoder, method and computer program using zero input response to obtain smooth transitions |
| US11170797B2 (en) | 2014-07-28 | 2021-11-09 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Audio decoder, method and computer program using a zero-input-response to obtain a smooth transition |
| US11922961B2 (en) | 2014-07-28 | 2024-03-05 | Fraunhofer-Gesellschaft Zur Foerderung Der Angewandten Forschung E.V. | Audio decoder, method and computer program using a zero-input-response to obtain a smooth transition |
| WO2019227931A1 (en) * | 2018-05-31 | 2019-12-05 | 华为技术有限公司 | Method and apparatus for calculating down-mixed signal |
| CN110556116A (en) * | 2018-05-31 | 2019-12-10 | 华为技术有限公司 | Method and apparatus for calculating downmix signal and residual signal |
| CN110556116B (en) * | 2018-05-31 | 2021-10-22 | 华为技术有限公司 | Method and apparatus for calculating downmix signal and residual signal |
| US11869517B2 (en) | 2018-05-31 | 2024-01-09 | Huawei Technologies Co., Ltd. | Downmixed signal calculation method and apparatus |
| US11961526B2 (en) | 2018-05-31 | 2024-04-16 | Huawei Technologies Co., Ltd. | Method and apparatus for calculating downmixed signal and residual signal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7568695B2 (en) | Harmonic Dependent Control of the Harmonic Filter Tool | |
| KR101023460B1 (en) | Signal Processing Method, Processing Unit and Voice Decoder | |
| EP2676270B1 (en) | Coding a portion of an audio signal using a transient detection and a quality result | |
| KR20020052191A (en) | Variable bit-rate celp coding of speech with phonetic classification | |
| US20130030798A1 (en) | Method and apparatus for audio coding and decoding | |
| CN107256709B (en) | Audio encoding device | |
| EP3000110A1 (en) | Apparatus and method for selecting one of a first encoding algorithm and a second encoding algorithm using harmonics reduction | |
| JP3628268B2 (en) | Acoustic signal encoding method, decoding method and apparatus, program, and recording medium | |
| WO2000041163A2 (en) | A method and apparatus for determining speech coding parameters | |
| CN101197134A (en) | Method and apparatus for eliminating influence of encoding mode switch-over, decoding method and device | |
| CN106409304B (en) | Time domain envelope processing method and device of audio signal and encoder | |
| JP2009530679A (en) | Method for post-processing a signal in an audio decoder | |
| EP3079151A1 (en) | Audio encoder and method for encoding an audio signal | |
| US20130096913A1 (en) | Method and apparatus for adaptive multi rate codec | |
| JP7608362B2 (en) | Method and device for detecting attacks in an audio signal to be coded and for coding the detected attacks - Patents.com | |
| Li et al. | A 1.8 kbps vocoder based on Mixed Excitation Linear Prediction | |
| CN1875401A (en) | Harmonic noise weighting in digital speech coders | |
| EP1521243A1 (en) | Speech coding method applying noise reduction by modifying the codebook gain | |
| CN102968997A (en) | Method and device for treatment after noise enhancement in broadband voice decoding | |
| JPH08160996A (en) | Voice encoding device | |
| CN116978391A (en) | Audio coding method, system, encoder, medium and equipment | |
| KR20000013870A (en) | Error frame handling method of a voice encoder using pitch prediction and voice encoding method using the same | |
| JPH09269798A (en) | Voice coding method and voice decoding method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20080611 |