TW200931396A - Apparatus for encoding and decoding audio signal and method thereof - Google Patents

Abstract

A method and/or apparatus for encoding and/or decoding an audio signal is disclosed, in which a downmix gain is applied to a downmix signal in an encoding apparatus which, in turn, transmits, to a decoding apparatus, a bitstream containing information as to the applied downmix gain. The decoding apparatus recovers the downmix signal, using the downmix gain information. A method and/or apparatus for encoding and/or decoding an audio signal is also disclosed, in which the encoding apparatus can apply an arbitrary downmix gain (ADG) to the downmix signal, and can transmit a bitstream containing information as to the applied ADG to the decoding apparatus. The decoding apparatus recovers the downmix signal, using the ADG information. A method and/or apparatus for encoding and/or decoding an audio signal is also disclosed, in which the method and/or apparatus can also vary the energy level of a specific channel, and can recover the varied energy level.

Description

200931396 VI. Description of the Invention: [Technical Field of the Invention] The present invention relates to a method and/or apparatus for encoding and/or decoding an audio signal. [Prior Art] The present invention relates to the encoding and/or decoding of spatial information of a multi-channel audio signal. In recent years, various encoding techniques and methods for digital audio signals have been developed, and various related products have been manufactured accordingly. However, when the audio signal of the Fuxi channel is downmixed in the form of a mono (m〇n〇) or a stereo (stere〇) audio signal, there may be a loss of the sound level of the audio signal (s〇und levd). In particular, since the encoded signal has a finite size, e.g., 16 bits, there is still a loss of sound level in the encoded signal even after encoding by the core codec. The above-mentioned sound drop phenomenon of the audio 影响 affects the output characteristic of the timbre and causes a decrease in sound quality. SUMMARY OF THE INVENTION In view of the above problems, the main purpose of the present invention is to reduce the sound level loss problem of multi-channel audio signals by applying a shrinking gain to the multi-channel audio job. Another object of the present invention is to solve the problem of loss of sound level of the singular channel sound carrier by applying - arbitrarily downmixing the gain to the multi-channel audio signal. A further object of the present invention is to solve the problem of loss of sound level of a multi-channel audio signal by applying a specific channel gain to a specific channel of the multi-channel audio signal. Another object of the present invention is to solve the problem of loss of sound level of multi-channel audio signals by using at least two of a downmix gain, an arbitrary downmix gain, and a specific channel gain. Therefore, in order to achieve the above advantages and in accordance with the purpose of the present invention, a method for decoding an audio signal disclosed by the present invention includes the steps of: separating a bit stream of a downmix signal from the audio signal; and applying a downmix gain to downmixing Signal to correct this downmix signal. In addition, in order to achieve the above advantages and in accordance with the purpose of the present invention, the present invention further provides a method for decoding an audio signal comprising the steps of: separating a downmix signal and a spatial information signal from a bit stream of the audio signal; The information signal converts the downmix signal to a multi-channel audio signal; and applies a downmix gain to the multi-channel audio signal. Furthermore, in order to achieve the above advantages and in accordance with the present invention, the present invention further provides an audio signal encoding method comprising the steps of: generating a downmix signal and a spatial information signal from a multi-channel audio signal; Apply a downmix gain to this downmix signal. In addition, in order to achieve the above advantages and in accordance with the purpose of the present invention, the present invention further provides an audio signal encoding method, comprising the steps of: applying a downmix gain to a multi-channel audio signal; and generating a downmix signal from the application. Multi-channel audio signal with this downmix gain. In order to achieve the above advantages and in accordance with the purpose of the present invention, the present invention further provides a decoding apparatus for a sound 200931396 frequency signal comprising: a demultiplexer, a separate-downmix signal and a spatial information signal self-audio signal bit. a streamer; a downmix gain application unit that applies a downmix gain to the downmix signal; and a multichannel generation unit that uses the spatial information signal to convert the downmix signal of the time downmix gain to _ Channel audio signal. Moreover, in order to achieve the above advantages and in accordance with the purpose of the present invention, the present invention provides an audio signal encoding apparatus comprising: - a button unit, which is generated by a multi-channel audio signal, a county signal, and a spatial information generating unit. Spatial information from the multi-channel audio signal, as well as -, the secret gain unit m secret gain to the downmix signal. [Embodiment] A preferred embodiment of the present invention is described in detail with reference to the drawings. "Figure 1" is a schematic diagram of the method used to enable humans to recognize spatial information of audio signals. The use of multi-channel audio signals is based on the fact that since humans recognize an audio signal three-dimensionally, audio signals can be represented in three-dimensional spatial information using a plurality of parameter settings. The spatial parameter 表示 representing the spatial information of the multi-channel audio signal 〃 includes channel level difference (CLD), inter-channel correlation (interchannel • coherence, ICC), and channel time difference (CTD). The channel can be 200931396. The step difference indicates the energy deviation between the two channels. The correlation between channels indicates the correlation between the two channels. The channel time difference represents the time difference between the two channels. Figure 1 shows how humans spatially identify audio signals and how the concept of spatial parameters is generated. Please refer to "1" "The direct sound wave from a remote source 101 reaches the left ear 107 of the person, and the other direct sound wave 1〇2 reaches the person's right ear after being circulated around the head. 106. Ο * Two direct sound waves 102 and 1〇3 are biased according to the arrival time and energy level. Due to the deviation of the scheme, the above-mentioned channel time difference and channel energy step difference parameters are generated. On the other hand, if the reflected sound waves 1 〇 4 and 1 〇 5 both reach the two ears of the person, or if the rice such as the remote sound source 101 contains a dispersed sound source, the sound waves having less correlation are observed in the human (4) ear. Therefore, the above channel_off parameter is generated. By using the spatial parameters generated according to the above principle, the new tone number can be transmitted in the form of a mono or a voice nickname, and the transmitted mono or physical voice signal can be outputted in the form of a voice channel. The present invention provides a method for correcting a mixed signal by using the spatial information when the mixed signal is converted into a multi-channel audio signal. Figure 2 is the loss of the audio level of the audio signal generated during the audio signal encoding process. . The loss of the audio signal is mainly caused by two factors. First, when the initial signal has a higher sound level, the above-mentioned level loss is generated. Second, when the number of input scales of the (4) downmix is also large, the above-mentioned order loss is generated. For example, compared with the case where three channels are converged into one channel, t-channels are more often caused to lose the volume loss when they are downmixed into one channel. The scale loss shown in Figure 2 corresponds to the downmixing of five channels into one channel reading. (4) The face of this hair is limited to the situation. The above-mentioned loss of sound level can be caused by various factors such as dipping. The section (8) of the "Fig. 2" section depicts the sound level of the initial signal from the five-sided group. Almost all ranges of a limited size (e.g., 16 bits) can be used for each channel of the initial signal. Part (b) of Figure 2 depicts the downmix signal produced by the downmix of the five channels. As shown in part (8) of Figure 2, the downmix signal can have multiple peaks that exceed the limit. Part (e) of “2®” shows the audio signal generated by the _ core codec (such as the Advanced Audio Frequency Codec (ACC c〇dec)) after encoding/decoding the downmix signal. Even in the case where the audio signal is generated in accordance with the encoding/decoding operation of the core code decimator, since the audio signal is present within a limited size (e.g., 16 bits), there may still be a loss of sound level. This loss of sound level can affect the output characteristics of multi-channel audio signals and result in a degradation in sound quality.第 FIG. 3 is a schematic diagram of a first encoding device according to an embodiment of the present invention, in which a downmixing gain is applied to a downmix signal to correct the downmix signal. The first encoding device includes a downmixing unit 302, a spatial information generating unit 303, a downmix gain applying unit 306, and a multiplexer 308. Referring to FIG. 3, the downmixing unit 302 downmixes a multi-channel audio signal 301 to generate a downmix signal 304. In "Picture 3", η" indicates the number of input channels. The downmix signal 304 can be a mono, stereo or multi-channel audio signal. 8 200931396 The spatial information generating unit 303 extracts spatial information from the multi-channel audio signal 301. Here, the spatial information 〃 indicates information about the audio signal channel, wherein the audio signal channel is used in the upmix downmix signal as a multi-channel audio signal, wherein the downmix signal is generated by downmixing the multi-channel audio signal. . The downmix gain application unit 306 applies a downmix gain to the downmix signal 304 to reduce the level of the downmix signal 304. Here, the collapsing gains indicate the application (eg, multiplication) to the value of the downmix signal or the multichannel audio signal to change the sound level of the downmix signal. In the 〇 encoding device, the application of the downmix gain to the downmix signal is mainly used to reduce the sound level of the downmix signal. For example, 'When a downmix gain greater than one is used, the 'downmix signal is multiplied by the reciprocal of the downmix gain' to reduce the total sound level of the downmix signal. A particular channel gain, such as low frequency gain (LFE) or surround gain, can be applied to at least one channel of the multi-channel audio signal 301. As described above, in the case where a particular channel gain has been applied to at least one of the channels of the multi-channel audio signal 301, the downmixing unit 302 can generate the downmix signal 3〇4 associated with the multi-channel audio signal 301. Then, the downmix gain is applied to the downmix signal 304. Of course, the downmix gain application unit 306 can apply the downmix gain in the process of generating the downmix signal 304 from the multichannel audio signal 301. The multiplexer 308 generates a bit stream 309, wherein the bit stream 309 includes a downmix signal 307 and a spatial information signal 305 to which the downmix gain has been applied. The spatial information signal 305 is composed of spatial information extracted from the spatial information generating unit 303. The bit stream 309r is transmitted to a decoding device. The bit stream 309 may also contain information about the downmix gain, 200931396, which is the downmix gain information. Fig. 4 is a schematic diagram of a first decoding apparatus according to an embodiment of the present invention. In the first decoding apparatus, a downmixing gain is applied to a downmix signal to correct the downmix signal. The first decoding device includes a demultiplexer 402, a downmix signal decoding unit 405, a spatial information signal decoding unit 406, a downmix gain application unit 409, and a multichannel generating unit 411. Referring to Fig. 4, the multiplexer 402 receives the bit stream 4〇1 of the audio signal and separates a coded downmix signal 403 and a coded spatial information signal 4〇4 from the bit stream 401. The downmix signal decoding unit 405 decodes the encoded downmix signal 403' and outputs the generated decoded signal to a downmix signal 407. The spatial information signal decoding unit 4〇6 decodes the encoded mother space information signal 404′ and outputs the resultant decoded signal as spatial information 4〇8. The downmix gain application unit 409 applies a downmix gain to the downmix signal 4〇7 to output a downmix signal 410 having an initial sound level. For example, when the downmix gain is greater than j, the downmix signal is multiplied by this downmix gain to increase its sound level. At the same time, the downmixing application unit 409 performs the application of downmixing gain in the process of converting the downmix signal into a multichannel audio signal. The multichannel generating unit 411 outputs the downmix signal 410 to which the downmixing gain is applied by using the spatial information 408 as a multichannel audio signal (out2). Fig. 5 is a schematic diagram of a second encoding apparatus according to an embodiment of the present invention. In the second encoding apparatus, a downmixing gain is applied to the multichannel audio signal to correct the 200931396 multichannel audio signal. Similar to the first encoding device, the second encoding device includes a downmixing unit 504, a spatial information generating unit 505, a downmix gain applying unit 502, and a multiplexer 508. As shown in "Fig. 5", the second encoding device is similar to the first encoding device. The second encoding device differs from the first encoding device in the position of the downmix gain application unit 502. That is, although the downmix gain is applied to the downmix signal in the first encoding device, the downmix gain is applied to the multichannel audio signal in the second encoding device. In more detail, the downmix gain application unit 502 applies a downmix gain to the multi-channel audio signal 501 to generate a multi-channel audio signal 503 to which the downmix gain is applied. The downmixing unit 504 downmixes the multi-channel audio signal 5〇3 to generate a downmix signal 506. The spatial information generating unit 505 extracts the spatial information from the multi-channel audio signal 503 to which the downmixing gain is applied. The multiplexer 508 generates a bit stream 509 comprising a downmix signal 5〇6 and a spatial information signal 507. ◎ "FIG. 6" is a schematic diagram of a second decoding apparatus according to the embodiment of the present invention. In the first decoding apparatus, the - downmixing surface should be an m channel audio signal to correct the multichannel audio signal. Similar to the first decoding device, the second decoding device includes a demultiplexer 602, a downmix signal decoding unit 6〇5, a spatial information signal decoding unit 606, a multichannel generating unit 609, and a downmixing gain application. Unit 611. Since the demultiplexer 6〇2, the downmix signal solution unit 0〇5, and the spatial information signal decoding unit 606 are the same or similar to the demultiplexer 4〇2 of the first decoding device shown in “Fig. 4” The downmix signal decoding unit 4〇5 and the spatial information signal decoding unit 11 200931396 406 ' are therefore not described in detail below. The multi-channel generation unit 609 uses the spatial information 608 to convert the downmix signal 607 into a multi-channel audio signal 610. The downmix gain application unit 611 applies a downmix gain to the multichannel audio signal 610 and thus outputs a multichannel audio signal (out2) to which the downmix gain is applied. When the decoding device cannot output a multi-channel audio signal by using spatial information, the downmix signal 607 can be directly output (outl) from the downmix signal decoding unit 605.第 FIG. 7 is a schematic diagram of a third encoding device according to an embodiment of the present invention. In the third encoding device, a downmixing gain is applied to a downmix signal to correct the downmix signal. The third encoding device includes a downmixing unit 702, a spatial information generating unit 703, a downmix gain determining unit 706, a downmix gain applying unit 708, and a multiplexer 710. Please refer to "Fig. 7". The third encoding device is similar to the first encoding device. The third encoding device is different from the first encoding device in that the third encoding device includes 〇, a downmix gain determining unit 706. The downmixing unit 702, the spatial information generating unit 703, the downmixing gain applying unit 708, and the multiplexer 710 and the first encoding device of the first encoding device shown in FIG. 3, the spatial mixing unit 302, and the downmixing The gain application unit 306 and the multiplexer 308 are therefore not described in detail below. The downmix gain determining unit 706 determines the downmix gain 707 to be applied to the downmix signal. The downmix gain determining unit 706 can determine the downmixing by measuring at least one of the sound level loss frequency and the sound level loss 12 200931396 when the multichannel audio signal 7〇1 is downmixed to generate a downmix signal 704. Gain. When it is assumed that "xk(n),' (k = 1, 2, 3, ..., N) represents each of the multichannel audio signals

N channel signal, and the downmix signal is generated as 〃 · ~ (") evening, downmix gain

The maximum value of N can be determined as "for example, when ai = l, a2 = l, a3 = 1, % = 1 / 10,000, % = 1 Λ ^ and know = 丨 / #, the maximum value of the downmix gain can be determined It is 4.73. When the maximum value of the downmix gain is rounded, it can be determined to be 4.第 FIG. 8 is a schematic diagram of a third decoding apparatus according to an embodiment of the present invention. In the third decoding apparatus, a downmixing gain is applied to a downmix signal to correct the downmix signal. The second decoding device comprises a demultiplexer 8〇2, a downmix signal decoding unit 8〇5, a spatial information signal decoding unit 8〇7, a downmix gain extraction unit 8〇8, and a downmix gain application. Unit 809 and a multi-channel generating unit 812. Referring to "Fig. 8", the third decoding device is similar to the first decoding device. The difference between the third third decoding device and the first decoding device is the downmix gain extracting unit 8〇8. The demultiplexer 802, the downmix signal decoding unit 8〇5, the spatial information signal decoding unit 807, the downmix gain application unit 809, and the multichannel generation unit 812 are the same or similar to the first shown in FIG. The decoding device demultiplexer 4〇2, the downmix signal decoding unit 405, the spatial information signal decoding unit 4〇6, the downmix gain application unit 4〇9, and the multichannel generating unit 411, so the details are not given below. Description. The downmix gain extraction unit 808 can extract the downmix gain information from a decoding space " signal number 804 or a decoded downmix signal 803. 13 200931396 "FIG. 9" is a schematic diagram of a bit stream containing downmix gain information according to an embodiment of the present invention. As shown in part (a) of Figure 9, the downmix gain information can be inserted into the space information signal 9〇2 of each frame, where the bit stream contains a downmix signal 9〇1 and a space. Information signal 902. As shown in part (b) of Figure 9, the downmix gain information can also be inserted into the downmix signal 903 of each frame. Also, the downmix gain information can be inserted into the bit stream of each of the multiple frames. For all boxes of a bit stream, the downmix gain can be a certain value, or 缩 for each box or every multiple box, the downmix gain is a change value. According to the 'executable-method' of the present invention, each frame or each of the frames of the spatial information signal has a header (or a construction information area), and the header includes downmix gain information. For each frame of the spatial information signal having a header, the decoding device extracts the downmix gain information from the header and applies the downmix gain to this block. On the other hand, for each of the plurality of frames of the spatial information signal having a header, the decoding device extracts the downmix gain information from the frame having the header. The decoding device then applies a downmix gain to the frame with headers and applies the downmix gain extracted from the previous header to the remaining blocks without the header. The header can be included in the frame of the spatial information signal periodically or non-periodically. The 'downmix gain information' can also be inserted into the header stream 904 of the bit stream as shown in section (c) of Figure 9. The header 904 contains construction information and the like. In this case, the downmix gain information can be inserted into the header as a separate value, or can be inserted into the header as a group value after grouping with other values, such as a particular channel gain. - In accordance with the present invention, another method can be performed in which the downmix gain information is inserted into the reserved area of the bit 200931396 7G stream without the use of additional bits. Furthermore, in accordance with the present invention, another method can be performed in which the method shown in part (b) and (c) of "9®" (a) is used. For example, as shown in (6) of Figure 9, the downmix gain is inserted into the header, and as shown in the @) section of Figure 9, the header can be inserted into the spatial information signal at the same time. In addition, the downmix gain can be inserted directly into the bit stream or selectively inserted into the bit stream in accordance with identification information as to whether or not the downmix gain is used. For example, the header of the bit stream may have a first © identification information as to whether or not to use the downmix gain. When the first identification information used based on the downmix gain is determined, each block of the bit stream has second identification information as to whether or not the downmix gain is used. When it is determined that the downmix gain should be used in a frame, the downmix gain is included in this frame. "10A" and "10B" are schematic diagrams of various types of downmix gains according to an embodiment of the present invention. The downmix gain can have various values. For example, the "1" and "10B" tables are composed of specific channel gains (such as surround gain and low frequency gain) and downmix gains. As shown in Table 1, M/sqrt(2)々 and 1/Sqrt (10G can be used for surround gain and low frequency gain respectively. For downmix gain, you can use eve or eve 1/2" 〇refer to Table 2, M/sqrt (2 wide and 々l/sqrt (10 wide can be used for surround gain and low frequency gain respectively. For downmix gain' can use "Γ, 1/2 々 or 々1/4". As shown in Table 3 〆l/sqrtp/ and l/sqrt(10)々 can be used for surround gain and low frequency gain respectively. For downmix gain, use "Γ, 夕 l/sqrt(2)〃 or " 1/2&quot 15 200931396 Refer to Table 4, " l/sqrt (2 wide and 1//sqrt (10 wide can be used for surround gain and low frequency gain respectively. For downmix gain, can be used, Γ, 夕 l/sqrt ( 2) 々,, 1 /2' or 夕 l/(2xsqrt(2) wide. Refer to Table 5, l/sqrt(2)々 and 夕 l/Sqrt(l〇)e for surround gain and low frequency gain, respectively. For the downmix gain, you can use 夕广, "3/4,,,, 2/3 or eve 1/2" as shown in Table 6, 夕 i/sqr^r and 〃 i/sqrtdO), Can be used for surround gain and low frequency gain respectively. For downmix gain, use "丨〃, 夕3/4〃,, 2/ 4" or eve 1/4". Although the surround gain and low frequency gain described in "Picture 10" and "Picture 1" are fixed to a specific value (for example, respectively "丨/叫枚〃 And the present invention is not limited to this. According to the present invention, as with the downmix gain, the surround gain and the low frequency gain can be selected from a plurality of specific values. According to the present invention, the deviation gain and the spread gain are selected. The external money can be used to determine the channel gain. "11th picture" is a method for preventing the sound quality around the frame from being lowered, wherein the sound quality is reduced by the approximate gain. ## As the sound level changes due to the downmix gain 'The drop in sound quality may occur around the sudden change in the pro-gain value. This is because the sudden change in the sound level occurs in the frame where the downmix gain value suddenly changes. Therefore, 'the setting is required—the transition period to cause the change in the downmix gain to occur. The influence can be stabilized. For this, the following process can be used for the stationary process. 200931396 DG(8)=adiPGWn-l) + (la(n)DGt(8), where n = 〇, l 2,, N. 'K can be the first in the above formula Column linear function or ordinary n-sequence polynomial function. When the downmix gain (DG) changes, a (a, can also be a function that can exhibit a smooth change such as a Gaussian function, a Hanning window function (hanningfUneti〇n) or Hamming function 〇 At the same time, despite the smoothing process described above, the adverse effects due to sudden changes in the downmix gain still exist. Therefore, a limitation can be performed in the encoding process ’ '(10) a sudden change in the stop-mix gain. Of course, even when the encoding device does not include a structure capable of preventing a sudden change in the downmixing gain, analysis for preventing a sudden change in the downmixing gain can be performed in the decoding device. For example, when using a downmix gain with a value of incremental or decremental change, the controllable downmix gain variable is located within an increment or decrement between successive boxes, or at each predetermined number of boxes (η box) One increment or decrement to prevent sudden downmix gain changes.第 FIG. 12 is a flow chart of an audio signal encoding method according to an embodiment of the present invention, wherein the encoding method applies a downmix gain to a downmix signal. Referring to Fig. 12, the encoding device performing the audio signal encoding method first receives a multi-channel audio signal (step 1201). Next, the downmixing unit of the encoding device downmixes the multichannel audio signal and then generates a downmix signal (step 1202). Although the downmix signal is obtained in accordance with the downmixing of the multichannel audio signal as described above, it is also possible to use a downmix signal directly input from the outside of the encoding device, such as an arbitrary downmix-signal. A spatial information signal is generated by a spatial information generating unit of the encoding device from a plurality of 2009 31396 channel audio signals (step 1202). Then, a downmix gain is applied to the downmix signal through the downmix gain application unit of the encoding device (step 1203). For example, when the downmix gain is greater than 1, the downmix signal is multiplied by the reciprocal of the downmix gain to reduce the level of the downmix signal. On the other hand, when the downmix signal is less than 1, the downmix signal is multiplied by this downmix gain to reduce the sound level of the downmix signal. Next, a bit stream including the condensed mixed signal and the spatial information signal to which the downmix gain is applied is generated by the multiplexer of the encoding device (step 1204). The generated bit stream can be transmitted to a decoding device (step 1204). The downmix gain can be applied to all boxes of the downmix signal of the bit stream. Although the method is preferably used for a downmix signal frame having a higher sound level, when the method is applied to a downmix signal having a low sound level, since a signal-to-noise ratio (SNR) may decrease, Will produce defects. Therefore, different downmix gain values can be used at predetermined time intervals. 〇 A downmix gain application syntax (syntax) can be defined in each box of the bit stream. In this case, the downmix gain can be selectively applied to each frame in accordance with the downmix gain application syntax. For example, applying the downmix gain to the downmix signal can be performed as follows. First, set a downmix gain in the header of the bit stream. At this point, the downmix gain can be applied to all boxes of the downmix signal affected by the header. Second, apply a separate downmix gain to each frame of the downmix * in accordance with a separately defined syntax. 18 200931396 The second 'uses a combination of the first method and the second method described above. That is, the downmix gain (hereinafter referred to as the first downmix gain) of all the blocks to be applied to the downmix signal is set. The first downmix gain can be used for all cycles or longer cycle ranges, such as 丨 to 2 seconds. Independent of the first downmix gain, the other downmix gain (hereinafter referred to as the second downmix gain) is applied to the downmix signal of each frame for gain control of the period not covered by the first downmix gain. When the decoded downmix signal is copied in the form of a mono or audible voice signal, as described above, the decoding of the downmix signal to which the downmix gain has been applied can be directly performed without considering the application to the downmix signal. The downmix gain. However, when the downmix signal is decoded and copied in the form of a multi-channel audio signal, the following method can be used. The first method applies a downmix gain to the full range of the downmix signal or to the range of the header of the downmix signal application to recover the sound level of the associated audio signal. The second method applies a shrink-mix signal to the downmix signal of each frame, or to a downmix signal that is smaller than a plurality of frames to which the header range is applied. ® The third method is a combination of the first method and the second method. That is to say, a downmix is applied to the downmix signal of each frame or every plurality of frames, and then another downmix gain is applied to all ranges of the downmix signal. Fig. 13 is a flow chart showing an audio signal decoding method according to an embodiment of the present invention, in which a downmix gain is applied to a downmix signal. Please refer to "Fig. 13" for the decoding device applying the audio signal decoding method to receive the bit stream of the audio signal (step 1301). The bit stream contains a coded downmix signal and a code space information 200931396 signal. The demultiplexer of the decoding device separates the encoded downmix signal and the encoded spatial information signal from the received bit stream (step 13〇2). The downmixed signal decoding unit of the gamma device decodes the encoded downmix signal and outputs a decoded downmix signal (step 1303). When the decanting device is unable to output a multi-channel audio signal by using the spatial information (step 1304), the decoding packet directly outputs the downmix signal decoded by the downmix signal decoding unit (step 1308). On the other hand, when the decoding device is capable of outputting the multi-channel audio signal ( (step 1304)', the following steps are performed. That is to say, the spatial information signal of the decoding device decodes the separated spatial information signal and generates spatial information. The downmix gain extraction unit of the decoding device extracts the downmix gain information from the spatial information signal or the downmix signal (step 1305). A downmix gain can be determined based on the extracted downmix gain information. The downmix gain application unit of the decoding device applies the determined downmix gain to the downmix signal (step 1306). The multichannel generating unit of the decoding device utilizes the spatial information to convert the downmix signal to which the downmix gain is applied to the multichannel tone signal (step 1307). Fig. 14 is a view showing an encoding apparatus according to an embodiment of the present invention, wherein an arbitrary downmix gain (ADG) is applied to a downmix signal to correct the downmix signal. The encoding device comprises a downmixing unit 1402, a spatial information generating unit 14〇3, a pitch mixing gain generating unit 1407, an arbitrary downmixing gain applying unit 14〇9 and a multiplexer 141b - please refer to "14th" The downmixing unit 1402 downmixes the multi-channel audio signals 14〇1, 200931396 to generate a downmix signal 1404. In "Figure 14," the number of input channels is indicated by Xi Xi. The spatial information generating unit 1403 extracts spatial information from the multi-channel audio signal 14〇ι. The arbitrary downmix gain generating unit 14〇7 can compare the downmix signal 1404 (hereinafter referred to as the first downmix signal) generated by the downmixing unit 14〇2 with the downmix signal 1405 directly input from the outside of the self-encoding device (hereinafter referred to as The second downmix signal) determines an arbitrary downmix gain. For example, the arbitrary downmix gain can be generated based on information indicating the difference between the first and second downmix signals 14〇4, 〇 1405, that is, the deviation information. Here, the arbitrary downmix gain " indicates the information used to reduce the difference between the second downmix signal and the first downmix signal. In the present invention, the 缩 arbitrary downmix gain 〃 can also be applied to the second downmix signal or the first downmix signal to correct the downmix signal. The arbitrary downmix gain application unit 1409 applies any downmix gain generated by any downmix gain generating unit 1407 to a downmix signal 14〇8. When the downmix signal 14〇8 is the second downmix signal 1405, the arbitrary downmix gain is not only used to reduce the second downmix signal.

The difference between the Q number 1405 and the first downmix signal 1404 is also used to correct the downmix signal 1408, such as reducing the sound level of the downmix signal 1408. In this case, any downmix gain applied to the downmix signal 1408 can be performed for each block. The multiplexer 1411 generates a bit stream 1412 containing the downmix signal 1410 and the spatial information signal 1406 to which any downmix gain has been applied. The spatial information signal 1406 is composed of spatial information extracted by the spatial information generating unit 1403. The bit stream 1412 is transmitted to a decoding device, and the bit stream 1412 can also contain information about any downmix gain. 21 200931396 "Figure 15" is a schematic diagram of a browning device according to an embodiment of the present invention, wherein - the arbitrary downmixing gain system is a bribe, and the positive interrogation I decoding device comprises - demultiplexing II 15G2 a downmix signal decoding unit, a spatial information signal decoding unit 1507, an arbitrary downmix gain extraction unit 15〇8, an arbitrary downmix gain application unit 1509, and a multichannel generation unit 1512. Please refer to "Fig. 15". The multiplexer 15〇2 separates a coded downmix signal 1503 and a coded space information signal 15〇4 from the bit stream 15〇1. The downmix signal decoding unit 1505 decodes the encoded downmix signal 15〇3, and outputs the generated decoded signal to the downmix signal 1506, wherein the downmix signal 1506 can be a mono, stereo or multi-channel audio signal. The downmix signal decoding unit 15〇5 can use a core codec. When the decoding device cannot process the downmix signal 15〇6 to output the multi-channel audio signal, the downmix signal 1506 can be directly output from the decoding device (〇utl) e spatial information signal decoding unit 1507 to decode the coded space information signal 1504, and output The generated decoded signal is spatial information 1511. The arbitrary downmix gain extraction unit 1508 extracts information about the arbitrary downmix gain from the spatial information signal 1504, that is, any downmix gain information. The arbitrary downmix gain extraction unit 1508 can also extract arbitrary downmix gain information from the downmix signal 1506. The arbitrary downmix gain application unit 1509 applies any downmix gain to the downmix signal 1506' where any downmix gain is determined in accordance with any downmix gain information extracted by any downmix gain extraction unit 15A8. The multi-channel generating unit 1512 uses the spatial information 1511 to convert the downmix signal 151 having an arbitrary downmixing gain into a multi-channel audio 22 200931396 signal 'and outputs a multi-channel audio signal (out2). Fig. 16 is a schematic view of the arrangement of the present invention, in which the encoding device cap gain and (4) the downmix gain wire are used for the test signal to correct the downmix signal. The encoding device includes a downmixing unit 16〇2, a spatial information generating unit 1603, a downmix gain applying unit 1606, an arbitrary downmix gain application unit_, and a multiplexer 1610. As shown in Figure 16, the downmixing unit 16〇2, the spatial information generating unit, the multiplexer and the multiplexer are the same or similar to the above-mentioned correspondences shown in Figure 14 and therefore are not given here. Detailed description. The coding device of "Fig. 16" is different from the coding device of "Fig. 14" in that the coding device of "Fig. 16" includes a downmix gain application unit 16〇6 and an arbitrary downmixing and thinning unit surface. An application that can perform relatives and arbitrary downmix gains. Although not shown in "Fig. 16," the encoding device of "Fig. 16" may include a downmix gain generating unit and an arbitrary downmix gain generating unit. In detail, the downmix gain unit is used to reduce the gain of the 1604 ° downmix signal to the full range of the downmix signal 1604. Moreover, the application of the downmix gain can be performed during the downmixing unit 1602 to downmix the multi-channel audio signal 1601 and thus generate the downmix signal 1604. 7 The arbitrary downmix gain application unit 16〇8 applies an arbitrary downmix gain to the downmix signal that has been applied with the downmix gain. As described above, applying any downmixing benefit to the downmix signal 1607 can be performed on each block. In accordance with the application of any downmix gain, the waveform of the downmix signal with 23 200931396 arbitrary downmix gain is applied, similar to the similar effect that occurs when dynamic range control (DRC) is applied. Any downmix gain can be applied to the downmix signal in the frequency domain, more specifically in the hybrid domain. According to the present invention, it is also possible to apply the downmix gain and the arbitrary downmix gain to the downmix signal input from the outside of the encoding device ("16th figure" is not shown). The multiplexer 1610 generates a bit stream 1611 containing a downmix signal 16〇9 and a spatial information signal 1605 applied with any downmix gain. Figure 17 is a schematic diagram of a decoding apparatus according to an embodiment of the present invention, wherein a downmixing gain and an arbitrary downmixing gain are applied to a downmix signal to correct a downmix signal. The decoding device comprises a demultiplexer 1702, a downmix signal decoding unit 17〇5, a spatial sub-signal decoding unit το HG7, a lag gain and an arbitrary downmix gain extracting unit 17〇8, and an arbitrary downmixing gain application. Unit 17〇9, a downmix gain application unit η. And a multi-channel generating unit 1714. ◎ Referring to "17th picture", the multiplexer 1702, the condensed code bribe unit 17〇5, the space 观 麟 num code unit (4), and the multi-channel generating unit (7) 4 have the same or similar correspondences as the "fifteenth figure" The component of Wei, shame this wealth and then give its detailed description. The difference between the decoding device of the "figure map" and the decoding device of the "figure 15" is that the decoding device of the "figure map" includes the pro-gain and arbitrary viewing gain extraction unit _, the arbitrary downmix gain application unit and The downmix gain should be ^Cui 1711 so the application of the downmix gain and any downmix gain can be performed. 24 200931396 , Ship gain and arbitrary downmix gain extraction unit ( from (4) information signal Π 04 extraction downmix gain _ intentional downmix gain. The downmix gain and any downmix gain information can be drawn through the side elements. Alternatively, the information can be extracted from individual components (not shown in Figure 17). In addition, the downmix gain and any downmix gain information may also be extracted from a downmix signal 1706. Arbitrary downmix gain application unit 17G9 application - arbitrary downmix gain to downmix signal 17G6 'its towel extension mixed gain Wei Zhao extracted arbitrary ship gain information generated, downmix signal 1706 according to the downmix signal decoding unit 17〇5 The decoding operation is generated. As described above, applying any downmix gain to the downmix signal 17〇6 can be performed for each frame. The downmix gain application unit 1711 applies the downmix gain generated in accordance with the downmix gain information to the downmix signal 1710, wherein the downmix signal 1710 has been applied with any downmix gain. The multichannel generating unit 1714 outputs the downmix signal 1712 to which the arbitrary downmixing gain and the downmixing gain have been applied, using the spatial information 1713, as a multichannel audio signal (〇ut2). When the field decoding device is unable to output such a multi-channel audio signal, the decoding device can directly rotate the downmix signal I7〇6 (〇utl) generated by the decoding operation of the downmix signal decoding unit 1705. A schematic diagram of a plurality of cheek bands with arbitrary downmixing gain applied to an embodiment of the invention. When applying any downmix gain to the frequency band of the audio signal, any downmix gain may have the same value as the channel energy level deviation (CLD) of the audio signal. For example, the 'arbitrary downmix gain can have the same number of parameter bands as the channel energy level deviation. 25 200931396 Therefore, when an arbitrary downmix gain is applied to the decoding apparatus, the number of groups can be determined in accordance with 'bsFreqResStridexxx' shown in Fig. 18, in which all frequency bands should be divided into this group. When, Stride, is i, grouping of all frequency bands is not performed. In this case, the reading of each fresh material downmix gain is performed, and the pitch-mixed gain is applied to the frequency band. When 〃pb(iv)e〃 is 5, the downmix gain is read every five bands and the arbitrary downmix gain of the read is applied to the five bands. On the other hand, when the pbstrif is 28, the reading of the arbitrary downmix gain is performed, and any pro-gain of the readout is applied to the money band. Therefore, when the lion is strong, the total band gain control is performed, and when the value of the gamma is not 则, the multi-band gain control is performed. It is also possible to perform the gain based on (4) the shrinkage of the job (4) Channel 〇 In addition, any reduction of the ylang also rises to the level. Here, the time slot represents a time interval through which the audio signal is evenly divided in the time domain. Therefore, when the sound level is facing the loud sound suddenly When the change occurs at a specific time position, the high-pitched gain control can be performed at this particular time position. When an arbitrary downmix gain value change occurs, an initial interpolation is performed on the arbitrary downmix gain. Otherwise, the scale Arbitrarily downmix the gain value. Therefore, in the case of the total band gain control, an arbitrary downmix gain for each time slot exists for all other bands. In the case of multi-band gain control, each time slot One of the 26 200931396 is intended to be used in multiple frequency bands. "19th picture" is a flowchart of the audio coding method of the present invention - an arbitrary secret gain Use the downmix signal to correct the downmix signal. The coding device that uses this sound ship rhyme code is first connected to the (4) Uyin Na (step 1901). Then, the squad number is shrunk by the scale element of the encoding device, and then the first downmix signal is generated (step 1902).产生 A spatial information signal is generated from the S channel audio signal through the empty service generating unit of the encoding device (step 19〇2). Then, the downmix signal directly input from the first downmix signal and the self-encoding device, that is, the second downmix signal, is compared by any downmix gain generating unit of the encoding device. According to the comparison result, the arbitrary downmix gain generating unit generates an arbitrary downmix gain (step 1903). Then, in any of the downmix signal application units of the encoding device, the generated arbitrary downmix gain is applied to the first downmix signal or the second downmix signal (step 1904) and then generated by the multiplexer of the encoding device. A bit stream with a downmix signal and a spatial information signal of any downmix gain is applied (step 19〇5). The generated bit stream is transmitted to a decoding device (step 19〇5). In accordance with the present invention, another audio signal encoding method can also be implemented in which the downmix gain and any downmix gain are applied to the downmix signal to correct the downmix signal. This encoding method is similar to the encoding method shown in Figure 19. The difference between the encoding method and the encoding method shown in Figure 19 is that the encoding method further includes applying the downmix gain to the reduction after generating the downmix signal and the spatial information signal as shown in "No. 19 27 200931396". The step of mixing the number. In this encoding method, then an arbitrary downmix gain can be applied to the downmix signal to which the downmix gain has been applied. According to the present invention, any downmix gain is generated in such a manner that the low frequency portion of any downmix gain is not generated as a gain, but is transmitted by performing residual coding of the low frequency component of the first downmix signal. The high frequency portion generated and, as in the conventional method, 'arbitrarily downmixed gain is generated as a gain such that the generated arbitrary downmix gain has an improved characteristic. Here, the residual code 〃 indicates the downmix signal of the directly coded portion. In the above method, the low frequency portion of any downmix gain is generated by performing direct residual coding for the low frequency component of the first downmix signal. However, the low frequency portion of any downmix gain can be generated by performing residual encoding for the difference between the first and second downmix signals. @ Any downmix gain generated as a gain and any downmix gain generated according to the low frequency component residual code of the first downmix signal is applied to the charm signal to correct the downmix signal. According to the present invention, the restoration information associated with the sound level loss generation point of the downmix signal can be added to any downmix gain, or can be transmitted along with any downmix gain, so that any downmix gain with restoration information can be used for the decoding device. Correction of the medium-mixed mixed signal. According to this (4), the information of the navigation domain correction (for example, (4) Airborne amplitude) and the information used to restore the second downmix signal to reduce the deviation between the second downmix signal and the first downmix signal 28 200931396, Can be included in an arbitrary downmix gain. The arbitrary downmixing gain generated according to the above manner may be included in the state in the spatial information signal transmitted. FIG. 20 is a flowchart of the audio signal decoding method according to an embodiment of the present invention, where an arbitrary downmix is used in the solution method. The gain (ADG) is applied to a downmix signal to correct the downmix signal. Referring to Fig. 20, the decoding device applying the audio signal decoding method receives a bit stream of an audio signal (step 2001). The bit stream includes a coded downmix signal and an encoded space information signal. The coded downmix signal and the coded spatial information signal are separated from the received bit stream by one of the encoding devices (step 2002). The separated downmix signal is kicked off by the downmix signal decoding unit of the decoding device (step 2〇〇3). When the decoding device can not output the downmix signal as a multi-channel audio signal by using the spatial information, the decoding device can directly output the decoded downmixed magnetic (pitch 2GG8) m through the downmix signal decoding unit. When the mixed signal is a multi-channel audio signal (step 2004), perform the following steps. That is to say, the separated spatial information signal is depleted by decoding the spatial information signal decoding unit of the split, thereby generating spatial information. Any of the downmix gain extraction units of the decoding device can also be extracted from the space (four) signal or downmix signal (step 2_. An arbitrary downmix gain can be determined based on the extracted arbitrary downmix gain information) Any of the downmix gain application units of the decoding device is determined to be a step-by-step (2) 6). According to the information of the space 29 200931396, the downmix signal applying the arbitrary downmix gain is converted into a multi-channel audio signal through the multi-channel generating unit of the decoding device, and the multi-channel audio signal is output from the decoding device (step 2007). Another decoding method can also be implemented in accordance with the present invention in which a downmix gain and an arbitrary downmix gain are applied to a downmix signal to correct the downmix signal. This decoding method is similar to the decoding method shown in Figure 20. The decoding method differs from the decoding method shown in FIG. 20 in that the method includes the step of applying a downmix gain to the downmix signal before applying any downmix gain to the downmix signal (step 2006). . The downmix gain information and any downmix gain information are extracted from a spatial information signal or a downmix signal by a downmix gain and an arbitrary downmix gain extraction unit (not shown). Then, the downmix gain generated based on the extracted downmix gain information is applied to the downmix signal. The downmix gain can be applied to the full range of downmix signals. Therefore, any downmix gain generated by 0 based on the extracted arbitrary downmix gain information is applied to this downmix signal. Applying any downmix gain to the downmix signal can be performed on each frame. Fig. 21 is a block diagram of an encoding apparatus for correcting an energy levd of a specific channel according to an embodiment of the present invention. The encoding device includes a specific channel level processing unit 2102, a downmixing unit 21〇4, a spatial information generating early element 2105' and a multiplexer 2108. Referring to "21st picture", the specific channel level processing unit 21〇2 receives a multi-channel audio signal 2101, corrects the 30 200931396 energy level of the specific channel of the received multi-channel audio signal 21〇1, and outputs the corrected multi-channel. Audio signal 2103. Here, the energy level represents a value proportional to the amplitude of the associated signal and includes a sound level. Whether the energy level of a particular channel changes and how it changes can be judged by measurement or calculation. Preferably, the energy level correction is achieved by applying a specific channel gain to a channel signal whose energy level changes. For example, by applying a surround gain or a low frequency (LFE) gain to a surround channel or a low frequency channel, this level correction can be achieved. The downmixing unit 2014 downmixes the multi-channel audio signal 2103 of the energy level correction, thereby generating a downmix signal 2106. Ο and the spatial information generating unit 2105 extracts spatial information from the multi-channel audio signal 2103. The multiplexer 2108 generates a bit stream 2109 containing a downmix signal 2106 and a spatial information signal 21〇7. The spatial information signal 2107 is composed of spatial information extracted by the spatial information generating unit 2105. The bit stream 21〇9 is transmitted to a decoding device, and the bit stream 2109 may also contain specific channel gain information. Fig. 22 is a block diagram of a decoding apparatus for correcting the energy level of a specific channel according to an embodiment of the present invention. The decoding device comprises a demultiplexer 2202, a downmix signal decoding unit 2205, a spatial information signal decoding unit 2206, a multichannel generating unit 2210, and a specific channel level processing unit 2212. Referring to Fig. 22, the multiplexer 2202 receives a bit stream 2201' of an audio signal and separates a coded downmix signal 22〇3 and a coded space information signal 2204 from the bit stream 2210. The downmix signal decoding unit 2205 decodes the encoded downmix signal 22〇3 and outputs a decoded downmix signal 2208 generated by the 31 200931396. The downmix signal decoding unit 2205 can also generate the downmix signal 2209 in the form of pulse code modulation (PCM) data by decoding the encoded downmix signal 2203'. The spatial information signal decoding unit 2206 decodes the spatial information signal 2204 and outputs the generated spatial information 2207. The multi-channel generating unit 2210 converts the downmix signal 2209 into a multi-channel audio signal 2211. The specific channel level processing unit 2212 receives the multi-channel audio signal 2211, the space 〇 2 signal 2207, and the downmix signal 2208, and performs energy level correction of each channel according to the received signal. The specific channel level processing unit 2212 includes a channel energy level detecting unit 2213', a correction determining unit 2214, and a channel level changing unit 2215. The channel level detection unit το 2213 detects whether and how the channel level of the multi-channel audio signal 2211 changes in each channel. The correction discriminating unit 2214 performs a shutdown based on the side of the channel energy level measurement ◎ early 70 2213. The shutdown-new correction is performed by each of the channels. The channel level correcting unit 2215 is based on the energy level of the peak positive-special path of the result of the discrimination performed by the correction determining unit 2214. When the standby decoding device is unable to rotate a multi-channel audio signal, the decoding device can directly output the downmix signal 22〇8 (outl), wherein the downmix signal is generated according to the decoding operation of the downmix signal unit 2205. On the other hand, when the decoding device is capable of outputting the channel audio signal, the decoding device can rotate the multi-sound audio signal (4) after correcting the energy level of the multi-channel audio signal of each channel. 32 200931396, When there is no energy level correction information about a specific channel issued by the self-encoding device, the decoding device shown in "Fig. 22" can correct the energy level of the specific channel by itself. This decoding apparatus has a feature that the specific channel level processing unit 2212 is independently constituted by the multichannel generating unit 2210. The channel-ordering unit 2213 included in the specific channel level processing unit 2212 can calculate the energy level of the initial audio signal according to the channel energy level deviation (CLD) and the downmix signal 2208 included in the spatial information. The calculated energy level is compared with the energy level ® of the multi-channel audio signal 2211 input from the multi-channel generating unit 2210. According to the comparison result, when the lying-in-the-edge correction is performed, the energy level correction operation is performed on the shed trimming unit 2215. Destroy Wei, the channel energy level correction, such as 5 multiplying the multi-channel audio signal, then the energy level is adjusted by a predetermined channel to further improve the multi-channel audio signal 2211. In this case, when there is a - energy level deviation, the correction unit 2214 may correct the channel energy level correction or only if the step deviation of the g existence exceeds a predetermined limit, the correction determination unit 2214 may determine that this must be performed. Channel level correction. According to this judgment, the other decoding device of the "decoding" of the "figure map" is different from the decoding device shown in the "nth diagram" in that the channel touch side unit and the correction unit It is included in the multi-channel generation unit' and the channel level correction unit is set up. According to the present invention, it is also possible to provide a decommissioning device for the decoding device of the "S 22 ®" hemp "but it is different from the decoding crack shown in the "Fig. 22" in 33 200931396 The side unit, the correction unit, and the unit 3 are generated in a single channel in the multi-channel. In this case, the decoding device ship can use the multi-channel generation of the internal function of the 単7C to perform the energy level correction of each channel. The energy level correction method using the internal function can contain the following method, that is, the # anal filter

(QMF) or hybrid filter rib adjustment of the upper arm gain method, used to adjust the total gain side (four) simple - ___ her) money moments. Chi) The method of value 'to adjust the subband envel〇pe application tool or time peak meenvebpe) _4 Wei Zhi green, when - go _ signal ', initial signal sum when 'used To adjust the above two signal gains, or to replace the above method, use the Green of the Terai group. #使肛镜相相器 (QMF) sister ageing n real __, simple material paving continuation characteristics. Field use-subband envd〇pe application of the king or time envelope (four) envelope) When the tool is short-coded, it can produce a final signal that provides realistic results. Fig. 23 is a view showing the aspect of the decoding apparatus according to an embodiment of the present invention for correcting the energy level of the specific money. This decoding device has a structure similar to that of the "figure" decoding device. Therefore, regarding the similar configuration, for example, the demultiplexer 2302 is reduced by the ring decoding unit 23〇5 and the spatial information signal decoding unit, the decoding device of the "23rd picture" will not be described in detail. The decoding device is located in the decoding device of 'the position of the channel level processing unit Μ(10). 34, 2009, 396, Please refer to FIG. 23, the specific channel energy level processing unit 2308 includes a channel energy level detecting unit 2309, a correction determining unit 2310, and a channel energy level correcting unit 2311. The specific channel level processing unit 2308 is capable of correcting the energy level of the downmix signal 2307 of each channel, wherein the downmix signal 2307 has a pulse code modulation (PCM) data format. In detail, when it is assumed that the energy level difference between the initial signal and the reproduced signal can be detected according to the energy level of the initial signal and the reproduced signal, the channel level correction unit 2311 corrects the downmix on a channel basis. The energy level of signal 2307. The specific channel level processing unit 2308 transmits a downmix signal 2312 to a multichannel generating unit 2313. After processing the downmix signal 2312 by a spatial information signal 2304, the multichannel generating unit 2313 can output the downmix signal 2312 as a multichannel audio signal 2314 (out2) 'where the space information signal decoding unit 23 〇 3 is used. The decoding operation of the spatial information signal generates spatial information in the spatial information signal 2304. Meanwhile, in accordance with the present invention, energy level correction of a special channel using a bit stream of the associated audio signal can be performed. In detail, when an encoding device corrects the energy level of a specific channel and transmits the correction information in a state in which the correction information is included in the one-bit stream, a decoding device receives the bit stream and can extract the correction information self-bit element. Streaming, and can restore the energy level of a particular channel based on the extracted correction information. For example, the encoding device sets a surround gain having a different value, in which a surround gain is selected to a surround frequency 'channel' and contains information about the applied surround gain, i.e., surround gain information in a bit 35 200931396 stream. At this time, the surround gain information can be included in the spatial information signal of the bit stream. The decoding device extracts surround gain information from the bit stream. Using this extraction information, the decoding device can restore the energy level of the surrounding channel to an initial level. In the following, a method of inserting correction information into a bit stream will be described in detail. First, format a spatial information signal so that each frame or each of the multiple frames has a header. Correction information about a particular channel (such as surround gain information) is included in this header. When each of the plurality of frames of the spatial information signal has a header, the header may be included in the spatial information signal periodically or non-periodically. Bitstreams can also contain bit information to indicate which channel should be amplified or attenuated' and how this channel is amplified or attenuated (decibel). In this case, the bit stream may contain information as to whether the energy level of a particular channel should be corrected, or whether the previous data should be used continuously when the correction is performed. The bitstream can also contain information about which channel should be corrected. In addition, the bitstream can contain information about the attenuation or amplification level (decibel) of the channel to be corrected. In accordance with the present invention, a method can be performed in a particular group of channels, and the adjustment of the characteristic channel gain can be performed for each group. In other words, in an encoding device, different channel gains are respectively differentiated from different specific channel groups. After the secret operation, the encoding device transmits the specific channel gain information in the state of the special channel gain information included in the _bit stream towel, and the towel bit stream is generated according to the cooperative operation*. A decoding device reciprocates the channel gain used by the encoding device to each group of multi-channel audio signals, thereby restoring the energy level of the multi-channel audio signal to an initial 36 200931396 energy level. For example, the channel of the audio signal can be divided into w, that is, the first group consists of a central channel, a left front channel, and a right touch channel, m left rear channel and a right rear channel, and - the third group consists of - low _ Road composition. In the above case, the channel-specific channel gain can be applied to each channel, and the resulting channel is summed to generate a mono downmix signal. In the decoding device, the mono __ is transmitted to a plurality of channels, Ο and each of the multi-channels of each group is multiplied by the side special singular gain, and then the mono downmix signal Qianyuan to - the performance is turned over ^ . The loading and landing gain multiplication can be performed after or during the transition step. - The first specific gain adjustment method can also be used. According to this second method, a specific channel gain is applied to each channel of each group. Therefore, the left front channel and the left rear channel are summed to generate a left channel, and the right front channel and the right rear channel are summed to generate a right channel. The characteristic channel gain is applied to each of the center channel and the low frequency channel, which is multiplied by 1/2 a (1/2) in turn. The resulting division is added to the left channel and the right channel's experience. When the generated live downmix signal is decoded and generated - the final channel gain, the specific channel gain should be performed in each group. In particular, the signals extracted from the left channel and the right channel of the downmix signal are multiplied by 2 Λ (1/2) and added to the center channel and the low frequency channel. Although the present embodiment described above relates to a mono or stereo signal, the present invention is not limited thereto. In accordance with the present invention, another method can also be practiced in which a _ downmix signal is generated after applying a particular channel 37 200931396 gain to each channel of each group, and an application of downmixing is performed for the resulting downmix signal. While the present invention has been described above in terms of the preferred embodiments thereof, it is not intended to limit the scope of the present invention, and it is possible to make some modifications and refinements without departing from the spirit and scope of the invention. The patent protection scope of the present invention is defined by the scope of the patent application attached to the specification. [Industrial Applications] As can be apparent from the above description, by applying a downmix gain to a downmix signal generated by downmixing of multichannel audio signals, or after applying a downmix gain to a multichannel audio signal, By downmixing the multi-channel audio signal, the present invention can effectively prevent the loss of sound level of a multi-channel audio signal. By applying an arbitrary downmix gain to a downmix signal generated by the downmixing of the multichannel audio signal, or after applying a downmix gain to the downmix signal, by applying an arbitrary downmix gain to the downmix signal, It can prevent the loss of sound level of multi-channel audio signal®. In addition, by correcting the energy level of the specific channel of the multi-channel audio signal and downmixing the modified multi-channel audio signal to generate a downmix signal, the sound level loss of the multi-channel audio signal can also be prevented. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a schematic diagram showing a method for enabling humans to recognize spatial information contained in an audio signal; 38 200931396 Fig. 2 is a waveform diagram of a sound level loss phenomenon occurring during audio signal encoding; A schematic diagram of a first encoding device according to an embodiment of the present invention, wherein a downmixing gain is applied to a downmix signal to correct the downmix signal; and FIG. 4 is a schematic diagram of a first decoding device according to an embodiment of the present invention; One of the downmix gains is applied to a downmix signal to correct the downmix signal. FIG. 5 is a schematic diagram of a second encoding device according to an embodiment of the present invention, wherein a downmix gain is applied to the multichannel audio signal. To modify the multi-channel audio signal; FIG. 6 is a schematic diagram of a second decoding device according to an embodiment of the present invention, wherein a downmix gain is applied to a multi-channel audio signal to correct a multi-channel audio signal; A schematic diagram of a third encoding device according to an embodiment of the present invention, wherein a downmixing gain is applied to a downmix signal to correct a downmix signal; FIG. 8 is a present invention A schematic diagram of a third decoding device in an embodiment, wherein a downmixing gain is applied to a downmix signal to correct the downmix signal; and FIG. 9 is a bit including a downmix gain information in an embodiment of the present invention, respectively. FIG. 10A and FIG. 10B are diagrams showing various types of downmix gains according to an embodiment of the present invention, and FIG. 11 is a method for preventing sound quality around the frame caused by the application of the downmix gain in the present invention; 12 is a flowchart of an audio signal encoding method according to an embodiment of the present invention. In the method of 39 200931396, the encoding method applies a downmixing gain to a downmix signal. FIG. 13 is a flowchart of an audio signal decoding method according to an embodiment of the present invention. FIG. 14 is a schematic diagram of an encoding apparatus according to an embodiment of the present invention, wherein an arbitrary downmix gain (ADG) is applied to a downmix signal to correct the reduction. FIG. 15 is a schematic diagram of a decoding apparatus according to an embodiment of the present invention, wherein an arbitrary downmix gain is applied to a downmix signal to correct a downmix signal; and FIG. 16 is an implementation of the present invention. The schematic diagram of the encoding device, wherein the downmix gain and any downmix gain are applied to the downmix signal to correct the downmix signal; FIG. 17 is a schematic diagram of the decoding device according to an embodiment of the present invention, wherein the downmix gain and the arbitrary downsizing The hybrid gain system is applied to the downmix signal to correct the downmix signal; FIG. 18 is a schematic diagram of a plurality of frequency bands to which any downmix gain is applied according to an embodiment of the present invention; and FIG. 19 is an audio signal according to an embodiment of the present invention; A flowchart of an encoding method, wherein an arbitrary downmixing gain is applied to the downmix signal to correct the downmix signal; and FIG. 20 is a flowchart of an audio signal decoding method according to an embodiment of the present invention, wherein the decoding method has an arbitrary downmixing gain Applying to a downmix signal to correct a downmix signal; FIG. 21 is a block diagram of an encoding apparatus according to an embodiment of the present invention, wherein the encoding device is used to correct a sound level of a specific channel; a block diagram of a decoding device for correcting a sound level of a specific channel; and 200931396 FIG. 23 is a block diagram of a decoding device according to an embodiment of the present invention This particular coding means for correcting a sound level of the channel. [Main component symbol description] 101 Remote sound source 102 Direct sound wave 103 Direct sound wave 104 Reflected sound wave

105 reflected sound wave 106 right ear 107 left ear 301 multi-channel audio signal 302 downmixing unit 303 spatial information generating unit 304 downmix signal 305 spatial information signal 306 downmix gain application unit 307 application of downmix signal 308 with downmix gain 309 bit stream 401 bit stream 402 demultiplexer 200931396 403 code downmix signal 404 code space information signal 405 downmix signal decoding unit 406 space information signal decoding unit 407 downmix signal 408 space information 409 downmix gain application unit

410 downmix signal 411 multi-channel generating unit 501 multi-channel audio signal 502 downmix gain application unit 503 multi-channel audio signal 504 with downmixing gain condensing unit 505 spatial information generating unit 506 downmix signal 507 spatial information signal 508 509 bit stream 601 bit stream 602 demultiplexer 603 code downmix signal 42 200931396 604 code space information signal 605 downmix signal decoding unit 606 spatial information signal decoding unit 607 downmix signal 608 space information 609 multi channel generation Unit 610 downmix signal

611 downmix gain application unit 701 multi-channel audio signal 702 downmix unit 703 spatial information generating unit 704 downmix signal 705 spatial information signal 706 downmix gain determining unit 707 downmix gain 708 downmix gain application unit 709 applied with downmix gain Downmix signal 710 multiplexer 711 bit stream 801 bit stream 802 demultiplexer 43 200931396 803 Decode downmix signal 804 Decode spatial information signal 805 Downmix signal decoding unit 806 Downmix signal 807 Space information signal decoding unit 808 Downmix gain extraction unit 809 downmix gain application unit

810 Downmix signal 811 Spatial information 812 Multi-channel generating unit 901 Downmix signal 902 Spatial information signal 903 Downmix signal 904 Header 1401 Multi-channel audio signal 1402 Downmixing unit 1403 Spatial information generating unit 1404 First downmix signal 1405 Second Downmix signal 1406 spatial information signal 1407 arbitrary downmix gain generation unit 44 200931396 1408 downmix signal 1409 arbitrary downmix gain application unit 1410 application of downmix signal with arbitrary downmix gain 1411 multiplexer 1412 bit stream 1501 bit stream 1502 solution multiplexer

1503 coded downmix signal 1504 coded space information signal 1505 downmix signal decoding unit 1506 downmix signal 1507 spatial information signal decoding unit 1508 arbitrary downmix gain extraction unit 1509 arbitrary downmix gain application unit 1510 application of downmixing with arbitrary downmix gain Signal 1511 Spatial Information 1512 Multi-channel generating unit 1601 Multi-channel audio signal 1602 Downmixing unit 1603 Spatial information generating unit 1604 Downmixing signal 45 200931396

1605 1606 1607 1608 1609 1610 1611 1701 1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 Spatial information signal downmix gain application unit Application of downmix signal with downmix gain Any downmix gain application unit with downmix gain and Any downmixing gain downmix signal multiplexer bit stream bit stream stream multiplexer code downmix signal code space information signal downmix signal decoding unit downmix signal space information signal decoding unit downmix gain and arbitrary downmix gain Extracting unit arbitrary downmix gain application unit applying downmixing signal with arbitrary downmixing gain downmixing gain application unit applying downmixing signal with arbitrary downmixing gain and downmixing gain spatial information multichannel generating unit 46 200931396 2101 multichannel audio Signal 2102 specific channel energy processing unit 2103 modified multi-channel audio signal 2104 downmix unit 2105 spatial information generating unit 2106 downmix signal 2107 spatial information signal

2108 multiplexer 2109 bit stream 2201 bit stream 2202 demultiplexer 2203 code downmix signal 2204 code space information signal 2205 downmix signal decoding unit 2206 space information signal decoding unit 2207 space information 2208 downmix signal 2209 downmix signal 2210 Multi-channel generating unit 2211 Multi-channel audio signal 2212 Specific channel level processing unit 47 1 200931396 2213 Channel level detecting unit 2214 Correcting unit 2215 Channel level correcting unit 2301 Bit stream 2302 Demultiplexer 2303 Spatial information signal Decoding unit 2304 spatial information signal

2305 Downmix signal decoding unit 2306 Downmix signal 2307 Downmix signal 2308 Specific channel energy processing unit 2309 Channel energy level detecting unit 2310 Correction unit 2311 Channel level correction unit 2312 Downmix signal 2313 Multichannel generating unit 2314 Multichannel Audio signal step 1201 Receive multi-channel audio signal Step 1202 Generate downmix signal and spatial information signal Step 1203 Apply downmix gain step 1204 Generate and transmit bit stream 48 200931396 Step 1301 Receive bit stream step 1302 Separate downmix signal and spatial information Signal step 1303 decodes the downmix signal step 1304 to output a multi-channel audio signal? Step 1305: Extracting the information about the downmix gain Step 1306 Apply the downmix gain step 1307 to output the multichannel audio signal Ο Step 1308 Output the downmix signal Step 1901 Receive the multichannel audio signal Step 1902 Generate the downmix signal and the spatial information signal Step 1903 Generate any Downmixing gain step 1904 applies arbitrary downmix gain to downmix signal step 1905 generates and transmits bit stream step 2001 receives bit stream ^ step 2002 separates downmix signal and spatial information signal step 2003 decodes downmix signal step 2004 outputs multichannel Audio signal? Step 2005 Extract information about any downmix gain Step 2006 Apply any downmix gain to the downmix signal Step 2007 Output multichannel audio signal Step 2008 Output downmix signal 49

Claims (1)

200931396 VII. Patent application scope: 1. A method for decoding an audio signal, comprising: receiving an audio signal, the audio signal comprising a downmix signal and a spatial information; and the downmix signal and the One of the spatial information extracts an arbitrary downmix gain information (ADGI); and applies an arbitrary downmix gain (ADG) based on the arbitrary downmix gain information captured. ) to the downmix signal. 2. The method of decoding an audio signal according to claim 1, wherein the arbitrary downmixing gain is applied to each frame of the downmix signal. 3. The method of decoding an audio signal according to claim 1, wherein the arbitrary downmixing gain is applied to each time slot of the downmix signal. 4. The method of decoding an audio signal according to claim 1, wherein the arbitrary downmix gain is independently applied to each frequency band of the downmix signal. 5. The method of decoding an audio signal according to claim 4, wherein the arbitrary downmix gain is independently applied to each time slot of the frequency band of the downmix signal. 6. The method of decoding an audio signal according to claim 1, further comprising: extracting the arbitrary downmix gain information from a header of a signal having the spatial information. 7. The method of decoding an audio signal according to claim 6, wherein the header is included in each box of the spatial information, or is included in each of the plurality of spatial information 50 200931396 boxes. 8. The method of decoding an audio signal according to claim 7, wherein the header is included in each of the plurality of frames of the spatial information periodically or non-periodically. 9. The method for decoding an audio signal according to claim 1, further comprising: extracting a downmix gain information (DGI) from one of the downmix signal and the spatial information; The downmix gain information applies a downmix gain to the downmix nickname. 10. The method of decoding an audio signal according to claim 9, wherein the downmix gain information is applied to all of the downmix signals, and the arbitrary downmix gain is applied to each frame of the downmix signal. 11. The method of decoding an audio signal according to claim 1, wherein the arbitrary downmix gain represents a comparison result between a first downmix signal and a second downmix signal. 12. The method of decoding an audio signal according to claim 11, wherein the first downmix code is encoded in an encoder, and the second downmix signal is provided by a device other than the encoder. A shuffling signal. 13. The method of decoding an audio signal according to claim 1, further comprising: decoding the arbitrary downmix gain information according to one of a plurality of predetermined values. 14. The method of decoding an audio signal according to claim 1, wherein the arbitrary downmix gain information is at least one bit, the bit being used to indicate whether the arbitrary downmix gain is applied. 51 200931396 • 15. The method of claim 16 wherein the ambiguous downmix gain beta comprises at least two bits, the two bits being used to refer to (4) using one of the arbitrary downmix gains degree. The method of decoding an audio signal as claimed in claim 15, wherein the degree is . 17. The method for decoding an audio signal according to claim 1, further comprising: omitting the arbitrary downmix gain information into a group consisting of a header of the audio signal, a frame of the downmix signal, and the spatial information. One of the groups. 〇18. A method for encoding an audio signal, comprising: generating a first downmix signal and a spatial information from a multi-channel audio signal; receiving a second downmix signal from an external source; applying an arbitrary downmix gain to An output downmix signal; and multiplexing the output downmix signal and the spatial information to generate the audio signal 'the audio signal has an arbitrary downmix gain information. 19. A decoder for decoding an audio signal, the audio signal being derived from a code; the decoding comprising: a demultiplexer configured to generate a downmix signal from the audio signal And an arbitrary downmix gain application unit, based on an arbitrary downmix gain information embedded in one of the downmix signal and the spatial information, the arbitrary downmix gain application unit is configured to downmix the arbitrary The gain is applied to the downmix signal. ^ 20. The decoder for decoding the audio signal according to claim 19, wherein the 52 200931396 arbitrary downmix gain system applies each pivot of the downmix signal. 21. The decoder for decoding the audio signal of claim 19, wherein the arbitrary downmix gain applies each time slot of the downmix signal. 22. The decoder for decoding the audio signal according to claim 19, further comprising: an arbitrary downmix gain information capturing unit configured to be from a header of a signal having the spatial information Take any arbitrary downmix gain information. 23. The decoder for decoding the audio signal as claimed in claim 22, wherein the 〇 header is included in each frame of a spatial information, or is included in each of a plurality of frames of spatial information. 24. The decoder for decoding the audio signal according to claim 19, further comprising: a downmix gain application unit based on a downmix gain information embedded in one of the downmix signal and the spatial information The downmix gain application unit is configured to apply the win-up gain information to a downmix gain to the downmix signal. 25. The decoder for decoding the audio signal according to claim 24, wherein the downmix gain information applies all of the downmix signal, and the arbitrary downmix gain system applies the downmix signal frame. 26. The decoder for decoding the audio signal of claim 19, wherein the arbitrary downmix gain represents a comparison between a first downmix signal and a second downmix signal. 27. The decoder for decoding the audio signal according to claim 26, wherein the first downmix signal is encoded in the encoder, and the second downmix signal is other than the 53 200931396 encoder. The downmix signal provided by a device. 28. The decoder for decoding the audio signal of claim 19, wherein the arbitrary downmix gain information encodes the arbitrary downmix gain information based on one of a plurality of predetermined values. 29. The decoder for decoding the audio signal of claim 19, wherein the arbitrary downmix gain information is at least one bit, the bit being used to indicate whether the arbitrary downmix gain is applied. The decoder for decoding the audio signal according to claim 19, wherein the arbitrary downmix gain information includes at least two bits, the two bits are used to indicate a degree of applying the arbitrary downmix gain . 31. The decoder for decoding the audio signal as recited in claim 30, wherein the degree is one. 32. The decoder for decoding the audio signal according to claim 19, wherein the arbitrary downmix gain information is embedded with a header of the audio signal, a Q frame of the downmix signal, and the spatial information. one. 33. An encoder for encoding an audio signal, comprising: a downmixing unit configured to generate a first downmix signal from a multi-channel audio signal; a spatial information generating unit configured to receive audio from the multi-channel A spatial information is generated in the signal; • an input port configured to receive a second downmix 54 from an external source 54 200931396; an arbitrary downmix gain application unit configured to apply an arbitrary downmix gain to And outputting the downmix signal; and a multiplexer configured to multiplex the output downmix signal and the spatial information, wherein an arbitrary downmix gain information is embedded in the output downmix signal and the spatial information. 55