KR100917845B1 - Apparatus and method for decoding multi-channel audio signal using cross-correlation - Google Patents

Abstract

The present invention relates to an apparatus and method for decoding a multichannel audio signal using cross-correlation. And a method for decoding a multichannel audio signal using cross-correlation for accurately reconstructing a center channel and a surround channel signal among the multichannel audio signals by adjusting the generated multichannel audio signal using sound source direction information). I would like to.

To this end, the present invention in the multi-channel audio signal decoding apparatus using cross-correlation, generating a multi-channel signal for generating a plurality of channel-specific audio signal from the downmixed stereo audio signal using the cross-correlation value between the left and right channels Way; And a cross-correlation value and a sub-band power value of the generated plurality of channel-specific audio signals and the sub-band cross-correlation information and the virtual sound source direction to restore the original signal for the downmixed stereo audio signal. Multi-channel signal adjusting means for adjusting using information.

Downmixing stereo audio signal, multichannel audio signal, surround left channel audio signal, surround right channel audio signal, cross correlation, virtual sound source direction information

Description

Multi-channel audio signal decoding apparatus using cross-correlation and its method {APPARATUS AND METHOD FOR DECODING MULTI-CHANNEL AUDIO SIGNAL USING CROSS-CORRELATION}

The present invention relates to an apparatus and method for decoding a multichannel audio signal using cross-correlation, and more particularly, to generate a multi-channel audio signal from a downmixed stereo audio signal using cross-correlation values between left and right channels and to encode encoded information (correlation). A multi-channel audio signal decoding apparatus using cross-correlation for accurately reconstructing the center channel and surround channel signals among the multi-channel audio signals by adjusting the generated multi-channel audio signals using correlation information and virtual sound source direction information; It's about how.

The present invention is derived from the research conducted as part of the next generation core technology development project of the Ministry of Information and Communication and the Ministry of Information and Communication Research and Development. [Task Management Number: 2005-S-403-02, Title: Intelligent Integrated Information Broadcasting ) Technology development].

With the recent popularization of home theater systems, the 5.1-channel audio format is becoming the mainstream of home audio. In addition, portable audio equipment has become a necessity to have a three-dimensional audio effect function that reproduces virtual surround by headphones or a small built-in speaker. This trend makes it possible to predict that the 5.1-channel audio format will be the default audio playback format for home and portable audio equipment.

However, the conventional 5.1 channel audio technology has a problem in that the amount of data increases with the number of channels. Therefore, in the conventional 5.1-channel audio technology, a multi-channel encoding method capable of compressing the data amount effectively performs an important function. For example, moving picture expert group (MPEG) -2 and MPEG-4 standardize the multi-channel coding method using the perceptual coding method. However, there is a problem in that the bit rate increases in proportion to the number of channels.

Recently, Binaural Cue Coding (BCC) schemes have been developed in which the bit rate does not increase even when the number of channels increases. The BCC is relatively simple in structure. After downmixing the multichannel audio to stereo or mono, a parameter for reconstructing the multichannel audio signal is calculated therefrom. These parameters may include Inter Channel Level Difference (ICLD), Inter Channel Time Difference (ICTD), and Inter Channel Cross-correlation (ICC).

In addition, Dolby Pro Logic is a representative technology for recovering a multi-channel audio signal from a stereo audio signal. However, in the case of Dolby Pro Logic, there is a problem that a signal that is unnecessarily removed or amplified in the spectrum may occur depending on the correlation between the stereo signals. In particular, when restoring a multi-channel audio signal from a stereo audio signal, there is a problem that the surround signal component is not correctly restored through simple addition and subtraction of the signal.

Therefore, in the prior art as described above, when restoring a multichannel audio signal that is an original signal from a downmixing stereo audio signal, the center channel, surround left channel, and surround right channel signal components are unnecessarily removed or amplified in the spectrum, thereby causing the center channel, And there is a problem that can not be faithfully restored to the surround channel signal component, it is an object of the present invention to solve this problem.

Accordingly, the present invention generates a multi-channel audio signal from the downmixed stereo audio signal using the cross-correlation value between the left and right channels and adjusts the generated multi-channel audio signal using encoding information (correlation information, virtual sound source direction information). Accordingly, an object of the present invention is to provide an apparatus and method for decoding a multichannel audio signal using cross-correlation for accurately reconstructing a center channel and a surround channel signal among multichannel audio signals.

The objects of the present invention are not limited to the above-mentioned objects, and other objects and advantages of the present invention which are not mentioned above can be understood by the following description, and will be more clearly understood by the embodiments of the present invention. Also, it will be readily appreciated that the objects and advantages of the present invention may be realized by the means and combinations thereof indicated in the claims.

The present invention has been proposed to solve the above problems, and generates a multi-channel audio signal from a downmixed stereo audio signal using cross-correlation values between left and right channels and uses encoding information (cross-correlation information, virtual sound source direction information). And adjusting the generated multi-channel audio signal.

More specifically, in the multi-channel audio signal decoding apparatus using cross-correlation, a multi-channel for generating a plurality of channel-specific audio signals from the downmixed stereo audio signal using the cross-correlation value between the left and right channels Signal generating means; And a cross-correlation value and a sub-band power value of the generated plurality of channel-specific audio signals and the sub-band cross-correlation information and the virtual sound source direction to restore the original signal for the downmixed stereo audio signal. Multi-channel signal adjusting means for adjusting using information.

Also, in the multi-channel audio signal decoding method using cross-correlation, a multi-channel signal generation step of generating a plurality of channel-specific audio signals from the downmixed stereo audio signal using cross-correlation values between left and right channels ; And a cross-correlation value and a sub-band power value of the generated plurality of channel-specific audio signals and the sub-band cross-correlation information and the virtual sound source direction to restore the original signal for the downmixed stereo audio signal. A multi-channel signal adjustment step of adjusting using the information.

As described above, the present invention generates multi-channel audio signals from downmixed stereo audio signals according to cross-correlation values between left and right channels, and uses multi-channel audio using spatial acoustic perception cues composed of cross-correlation between channel and virtual sound source direction information. By adjusting the signal, it is possible to accurately restore the center channel and surround channel signals among the multichannel audio signals.

In addition, the present invention has the effect of mitigating the spectral distortion phenomenon by adjusting the multi-channel audio signal using the spatial acoustic perception terminal composed of the cross-correlation between the channel and the virtual sound source direction information.

The above objects, features, and advantages will become more apparent from the detailed description given hereinafter with reference to the accompanying drawings, and accordingly, those skilled in the art to which the present invention pertains may share the technical idea of the present invention. It will be easy to implement. In addition, in describing the present invention, when it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a configuration diagram of an apparatus for decoding a multi-channel audio signal using cross-correlation according to the present invention, which is illustrated together with the apparatus for encoding a multi-channel audio signal.

As shown in FIG. 1, the multi-channel audio signal decoding apparatus 120 using cross-correlation according to the present invention includes a multi-channel signal generator 121 and a multi-channel signal adjuster 122, which are encoded by reference. The device 110 includes a downmixing unit 112 and a spatial acoustic perceptual cue analysis unit 111.

Hereinafter, each component of the encoding apparatus 110 and the decoding apparatus 120 will be described in detail.

The spatial acoustic perceptual analysis unit 111 receives the multichannel audio signal and subband filters the multichannel audio signal for each subband associated with each channel. In addition, the spatial acoustic perceptual analysis unit 111 extracts the spatial acoustic perceptual cue by analyzing the level difference and cross-correlation between adjacent channels in the audio signal of each subband filtered channel. Here, the spatial acoustic perception cues include cross-correlation values between the channels and virtual sound source direction information.

The downmixing unit 112 performs a function of compressing the multi-channel audio signal received from the spatial acoustic perception cue analysis unit 111 into a stereo audio signal. That is, the downmixing unit 112 mixes the subband filtered multi-channel sound spectrum by the spatial acoustic perceptual analysis unit 111 into a downmixing stereo audio signal and converts the downmixing stereo audio signal into a time domain signal.

At this time, the formula of the general matrix for downmixing in the downmixing unit 112 is shown in Equation 1 below.

L _dm = L + SL + SQRT (2) / 2 × C

R _dm = R + SR + SQRT (2) / 2 x C

Where L _dm And R _dm are the downmix left channel and downmix right channel stereo signals, L and R are the left and right channel signals in the multichannel sound signal, SL and SR are the surround left and surround right channel signals, and C is the center. Indicates a channel signal. A low frequency effect (LFE) signal, which is commonly used, can be processed by dividing it into both downmixing channels as in C.

Hereinafter, each component of the multi-channel audio signal decoding apparatus 120 using cross correlation according to the present invention will be described in detail.

The multichannel signal generator 121 separates the downmixing stereo audio signal in which the multichannel audio signal (the original signal) is downmixed by the encoding apparatus 110 according to the cross-correlation value between the respective channels to generate the downmixing multichannel audio signal. Create The multi-channel signal generator 121 adaptively extracts a common signal and an independent signal between the two channels by using an adaptive filter from the received downmixed stereo audio signal and performs a function of partially separating the surround channel signal and the center channel signal. do. That is, the multi-channel signal generation unit 121 uses the adaptive filter and the sum and difference signals between the two channels from the downmixing stereo (left channel and right channel) audio signals of the encoding device 110 to use the downmixing central channel signal. And generate a downmix surround channel signal.

In addition, the multi-channel signal adjusting unit 122 converts the channel cross-correlation value of the downmixing multi-channel audio signal generated by the multi-channel signal generating unit 121 to the inter-channel cross-correlation information of the original signal received from the encoding apparatus 110. The power value for each subband of the adjusted multichannel audio signal is adjusted according to the virtual sound source direction information of the original signal received from the encoding apparatus 110. The multi-channel signal adjusting unit 122 uses sub-band spectra of the downmixing multi-channel sound signal generated by the multi-channel signal generator 121 by using the cross-correlation and virtual sound source direction information extracted from the encoding apparatus 110. Adjust cross-correlation and shape of. That is, the multi-channel signal adjusting unit 122 adjusts cross-correlation and subband power of the multi-channel sound signal output from the multi-channel signal generating unit 121 to restore the original multi-channel signal. The multi-channel signal adjusting unit 122 outputs a multi-channel audio signal whose cross-correlation and shape are adjusted.

2 is a detailed configuration diagram of an embodiment of the spatial acoustic perception cue analysis unit of FIG. 1 used in the present invention.

As shown in FIG. 2, the spatial acoustic perceptual cue analysis unit 111 includes the first to fifth subband filtering units 201 to 205 and the spatial acoustic perceptual extracting unit 206 corresponding to each channel. Include.

The first to fifth subband filtering units 201 to 205 classify the multichannel audio signal input from the outside into subbands based on human auditory characteristics for each channel. The first to fifth subband filtering units 201 to 205 transmit the subband filtered first to fifth channel audio signals to the spatial acoustic perceptual end extractor 206.

The spatial acoustic perceptual extractor 206 analyzes the first to fifth channel audio signals subband filtered by the first to fifth subband filtering units 201 to 205, respectively, and correlates information between adjacent channels. Spatial acoustic perception cues containing virtual sound source direction information are extracted. That is, the spatial acoustic perceptual extractor 206 generates cross-correlation information and virtual sound source direction information between channels for each subband. The spatial acoustic perceptual extractor 206 transmits the first to fifth channel audio signals to the downmixer 112, and transmits the generated cross-correlation and virtual sound source direction information to the decoding device 120. do.

Here, the cross-correlation information between channels may be calculated in the frequency domain for each subband signal. In addition, the virtual sound source direction information may be calculated as an angle value between the adjacent channel speaker placement angles by the subband power ratio of the adjacent channel signal.

3 is a detailed block diagram of an embodiment of the multi-channel signal generator of FIG. 1 according to the present invention.

As shown in FIG. 3, the multi-channel signal generator 121 includes a first surround channel signal generator 310, a second surround channel signal generator 320, and a center channel signal generator 330. do. Here, the first surround channel signal generator 310 includes a first adaptive filter 311 and first and second subtractors 312 and 313. In addition, the second surround channel signal generator 320 includes a second adaptive filter 321 and third and fourth subtractors 322 and 323. In addition, the central channel signal generator 330 includes an adder 331 and a divider 332.

The multichannel signal generator 121 separates the downmixing stereo audio signal in which the multichannel audio signal (the original signal) is downmixed by the encoding apparatus 110 according to the cross-correlation value between the respective channels to generate the downmixing multichannel audio signal. Create Here, the surround channel signal component of the downmixing multichannel audio signal is obtained by using a difference value between the downmixing left channel signal and the downmixing right channel signal to update the coefficients of the first and second adaptive filters 311 and 321. The multi-channel signal generator 121 may remove a phenomenon in which a specific spectrum signal is distorted due to a phase difference by a smoothing effect using an adaptive filter.

Hereinafter, the components of the multi-channel signal generator 121 will be described in detail.

The first surround channel signal generator 310 removes the center channel signal component and the surround right channel signal component from the downmixing left channel audio signal among the downmixing stereo audio signals to remove the downmixing surround left channel signal. Create That is, the first surround channel signal generator 310 receives the downmixing right channel signal and the downmixing left channel signal, and the first adaptive filter 311 and the first and second subtractors 312 and 313 from the input signal. To generate the downmix surround left channel signal.

Here, the first adaptive filter 311 performs a function of suppressing a central channel signal component which is a common signal component and passing a surround signal which is an independent signal component. The first subtractor 312 outputs an error signal by subtracting the downmixing left channel signal passed through the first adaptive filter from the downmixing right channel signal. In this case, the output error signal is used to update the coefficient of the first adaptive filter 311. The second subtractor 313 subtracts the output signal of the first subtracter 312 from the downmix left channel signal to generate a downmix surround left channel signal. Here, subtracting the output signal of the first subtractor 312 from the downmix surround left channel signal is for maximizing forward and backward cross-correlation.

The second surround channel signal generator 320 removes the center channel signal component and the surround left channel signal component from the downmixing right channel audio signal among the downmixing stereo audio signals to remove the downmixing surround right channel signal. Create That is, the second surround channel signal generator 320 receives the downmixing left channel signal and the downmixing right channel signal, and the second adaptive filter 321 and the third and fourth subtractors 322 and 323 from the input signal. To generate the downmix surround right channel signal.

Here, the second adaptive filter 321 suppresses the central channel signal component that is a common signal component and passes the surround signal that is an independent signal component. The third subtractor 322 subtracts the downmixing right channel signal from the downmixing left channel signal to output an error signal. In this case, the output error signal is used to update the coefficient of the second adaptive filter 321. The fourth subtractor 323 subtracts the output signal of the second subtractor 322 from the downmixing right channel signal to output the downmixing surround right channel signal. Here, subtracting the output signal of the third subtractor 322 from the downmix surround right channel signal is for maximizing forward and backward cross-correlation.

The center channel signal generator 330 generates a downmixed center channel signal by combining a left channel audio signal and a right channel audio signal among the downmixed stereo audio signals. That is, the center channel signal generator 330 receives the downmixing left channel signal and the downmixing right channel signal, adds the downmixing two channel signals, and divides them in half to generate the downmixing central channel signal.

4 is a detailed configuration diagram of an embodiment of the multi-channel signal adjusting unit of FIG. 1 according to the present invention.

As shown in FIG. 4, the multi-channel signal adjusting unit 122 calculates the sixth to tenth subband filtering units 401 to 405, the first and second cross-correlation adjusting units 406 and 407, and the multichannel power ratio. A unit 408 and a signal converter 409 are included.

The multi-channel signal adjusting unit 122 adjusts the cross-correlation value of the channels of the downmixing multi-channel audio signal generated by the multi-channel signal generating unit 121 according to the cross-correlation information of the channels of the original signal, and adjusts the adjusted multi-channel. The power value of each subband of the audio signal is adjusted according to the virtual sound source direction information of the original signal. That is, the multichannel signal adjuster 122 adjusts the cross-correlation and subband power of the multichannel sound signal generated by the multichannel signal generator 121 to perform a function of restoring the original multichannel audio signal.

Hereinafter, each component of the multi-channel signal adjusting unit 122 will be described in detail.

The sixth to tenth subband filtering units 401 to 405 each perform subband filtering on the downmixing multichannel audio signal generated by the multichannel signal generator 121.

The multi-channel power ratio calculator 408 calculates the power ratio for each subband of the multi-channel signal from the virtual sound source direction information received from the encoding apparatus 110.

In addition, the first and second cross-correlation adjustment units 406 and 407 may use the original signal as the cross-correlation value of the channel of the downmixed multichannel audio signal filtered by the sixth to tenth subband filtering units 401 to 405, respectively. Adjust the channel's cross-correlation information.

In addition, the signal converter 409 calculates, by the multi-channel power ratio calculator 408, a subband power value of the multi-channel audio signal whose cross-correlation is adjusted by the first and second cross-correlation adjustment units 406 and 407, respectively. It adjusts to the multichannel power ratio which is made and converts it to time domain. That is, the signal converter 409 is a sub corresponding to the power of the surround signal output by the first and second cross-correlation adjustment units 406 and 407 according to the power ratio calculated by the multi-channel power ratio calculator 408. The power value of the multi-channel sound signal is adjusted for each band, and the signal whose power value is adjusted is converted into the time domain.

5 is a flowchart illustrating a method of decoding a multichannel audio signal using cross-correlation according to the present invention.

First, referring to the encoding method, the spatial acoustic perceptual analysis unit 111 subband-filters a multi-channel audio signal and includes spatial correlation perceptual information including cross-correlation and virtual sound source direction information between adjacent channels from each filtered channel audio signal. Extract

The downmixer 112 downmixes each filtered channel audio signal into a stereo audio signal and encodes the stereo audio signal.

Hereinafter, a multichannel audio signal decoding method according to the present invention will be described.

The multi-channel signal generator 121 separates the down-mixed stereo audio signal from which the multi-channel audio signal (the original signal) is downmixed from the down-mixed stereo signal received from the encoding apparatus 110 according to the cross-correlation value between the channels and downmixes them. A multichannel audio signal is generated (502).

The multi-channel signal adjusting unit 122 adjusts the cross-correlation value of the generated downmixing multi-channel audio signal according to the cross-correlation information of the original signal (504).

Thereafter, the multi-channel signal adjusting unit 122 adjusts and decodes the power value of each subband of the adjusted multi-channel audio signal according to the virtual sound source direction information of the original signal (506).

On the other hand, the method of the present invention as described above can be written in a computer program. And the code and code segments constituting the program can be easily inferred by a computer programmer in the art. In addition, the written program is stored in a computer-readable recording medium (information storage medium), and read and executed by a computer to implement the method of the present invention. The recording medium may include any type of computer readable recording medium.

The present invention described above is capable of various substitutions, modifications, and changes without departing from the technical spirit of the present invention for those skilled in the art to which the present invention pertains. It is not limited by the drawings.

1 is a configuration diagram of an apparatus for decoding a multichannel audio signal using cross-correlation according to the present invention;

2 is a detailed configuration diagram of an embodiment of the spatial acoustic perception cue analysis unit of FIG. 1 used in the present invention;

3 is a detailed configuration diagram of an embodiment of the multi-channel signal generator of FIG. 1 according to the present invention;

4 is a detailed configuration diagram of an embodiment of the multi-channel signal adjusting unit of FIG. 1 according to the present invention;

5 is a flowchart illustrating a method of decoding a multichannel audio signal using cross-correlation according to the present invention.

* Explanation of symbols on the main parts of the drawing

120: decoding apparatus 121: multi-channel signal generation unit

122: multi-channel signal adjusting unit 310: first surround channel signal generating unit

320: second surround channel signal generator 311: first adaptive filter

321: second adaptive filter 330: center channel signal generator

401 to 405: sixth to tenth subband filtering units

406: first cross-correlation adjustment unit 407: second cross-correlation adjustment unit

408: multi-channel power ratio calculator 409: signal converter

Claims (8)

In the multi-channel audio signal decoding apparatus using cross-correlation, Multi-channel signal generating means for generating a plurality of channel-specific audio signals from the downmixing stereo audio signal using left / right channel correlation values; And The cross-correlation value and the sub-band power value of the generated plurality of channel-specific audio signals and the sub-band power correlation information may be used to restore the original signal with respect to the downmixed stereo audio signal. Means for adjusting multichannel signals Multi-channel audio signal decoding apparatus using a cross-correlation comprising a. The method of claim 1, The multi-channel signal generating means, First surround channel signal generation means for generating a downmix surround left channel signal by removing a center channel signal component and a surround right channel signal component by using a cross-correlation value from a downmixing left channel audio signal among the downmixing stereo audio signals ; Generating a second surround channel signal for generating a downmix surround right channel signal by removing a center channel signal component and a surround left channel signal component by using a cross-correlation value from a downmixing right channel audio signal among the downmixing stereo audio signals Way; And Center channel signal generation means for generating a downmix center channel signal by combining a left channel audio signal and a right channel audio signal among the downmixed stereo audio signals Multi-channel audio signal decoding apparatus using a cross-correlation comprising a. The method of claim 2, The first surround channel signal generating means, And an adaptive filter and a subtractor for removing the center channel signal component and the surround right channel signal component by using the cross-correlation value from the downmixing left channel audio signal. The method of claim 2, The second surround channel signal generating means, And an adaptive filter and a subtractor for removing the center channel signal component and the surround left channel signal component by using the cross-correlation value from the downmixing right-channel audio signal. The method according to any one of claims 1 to 4, The multi-channel signal adjusting means, Multichannel power ratio calculating means for calculating a multichannel power ratio from the virtual sound source direction information received from the encoding apparatus; A plurality of subband filtering means for subband filtering each of the downmixing multichannel audio signals generated by the multichannel signal generating means; A plurality of cross-correlation adjustment means for adjusting the cross-correlation value of the subband filtered downmixing multichannel audio signal according to the cross-correlation information of the original signal; And Signal conversion means for adjusting the power value of each subband of the multi-channel audio signal whose cross-correlation is adjusted according to the calculated multi-channel power ratio and converting it into a time domain Multi-channel audio signal decoding apparatus using a cross-correlation comprising a. In the multi-channel audio signal decoding method using cross-correlation, Generating a plurality of channel-specific audio signals from the downmixing stereo audio signal using the cross-correlation value between the left and right channels; And The cross-correlation value and the sub-band power value of the generated plurality of channel-specific audio signals and the sub-band power correlation information may be used to restore the original signal with respect to the downmixed stereo audio signal. Multi-channel signal adjustment step Multi-channel audio signal decoding method using cross-correlation comprising a. The method of claim 6, The multi-channel signal generation step, Generating a downmixing surround left channel signal by removing a center channel signal component and a surround right channel component using a cross-correlation value from a downmixing left channel audio signal among the downmixing stereo audio signals; Generating a downmix surround right channel signal by removing a center channel signal component and a surround left channel signal component using a cross-correlation value from a downmixing right channel audio signal among the downmixing stereo audio signals; And Combining a left channel audio signal and a right channel audio signal among the downmixing stereo audio signals to generate a downmix center channel signal; Multi-channel audio signal decoding method using cross-correlation comprising a. The method according to claim 6 or 7, The multi-channel signal adjustment step, A multichannel power ratio calculating step of calculating a multichannel power ratio from the virtual sound source direction information received from the encoding apparatus; A subband filtering step of subband filtering each of the downmixing multichannel audio signals generated in the multichannel signal generation step; A cross-correlation adjustment step of adjusting the cross-correlation value of the subband-filtered downmixing multichannel audio signal according to the cross-correlation information of the original signal; And A signal conversion step of adjusting the power value of each subband of the multi-channel audio signal whose cross-correlation is adjusted according to the calculated multi-channel power ratio and converting it to the time domain Multi-channel audio signal decoding method using cross-correlation comprising a.

Images