JP6510870B2 - Program sound channel number conversion device, broadcast program reception device, and program sound channel number conversion program - Google Patents

Description

  The present invention relates to a program audio channel number converter for converting the number of channels of program audio in order to reproduce the sound of a different number of channels from that of the production of program audio at the time of transmission or reception of program audio. Apparatus and program audio channel number conversion program

  Conventionally, when converting the number of channels of program audio, conversion coefficients set in advance (see non-patent documents 1 and 2) or positions assumed when reproducing audio signals of respective channels (positions of assumed speakers And transform coefficients (see Non-Patent Document 3) calculated from angles. Specifically, the audio signal of each channel before channel number conversion is multiplied by the conversion coefficient, and the multiplication result is added to generate an audio signal after channel number conversion.

  For example, the technology for converting the number of channels from 22.2 channel acoustic system to 5.1 channel audio system and the technology for converting the number of channels from 5.1 channel audio system to 2 channels acoustic system are described in Non-Patent Document 1 Recommendation ITU-R BS.2 of the standard ARIB STD-B 32 and the aforementioned Non-Patent Document 2. 775-3 and the like.

  Such a channel number conversion method is based on the reproduction position of the audio signal of each channel, and a constant conversion coefficient is used regardless of whether the audio signal is a dialog, an instrumental sound or a background sound. .

  However, for example, when the viewer of the program is an elderly person, there is a tendency that the dialog is difficult to hear due to the influence of the background sound. In this case, it is desirable that the sound pressure levels of the dialog and the background sound be adjusted to be suitable for elderly people (see Non-Patent Document 4).

  On the other hand, there is known a measurement method of loudness value for evaluating the size of program sound, and management of program sound level is performed by the average loudness value measured by this loudness measurement method (non-patent document) 5). According to this loudness measurement method, the average loudness value of all sections of the program (all time sections from the start time point to the end time point of the program) falls within the tolerance (± 1 dB) of the target loudness value (-24 LKFS) The operation of program production is defined (see Non-Patent Document 6). This makes it possible to unify the volume feeling of program sound between programs and between broadcast stations.

ARIB Standard, ARIB STD-B32 Version 3.0, "Video Coding, Audio Coding and Multiplexing in Digital Broadcasting" Recommendation ITU Recommendation ITU-R BS. 775-3, "Multichannel stereophonic sound system with and without accompanied picture" Ville Pulkki, “Localization of amplitude-panned virtual sources II: two- and three-dimensional panning”, J. Audio Eng. Soc., Vol. 49, No. 9, pp. 753-767, 2001 Komori, Miyako, Kurozumi, Shibata, Murakawa, "Technology volume objective evaluation technology for broadcast programs easy to hear even by the elderly," IEICE technical research report W.I.T., no.164, pp.35-40 , 2010 Recommendation ITU Recommendation ITU-R BS. 1770-3, "Algorithms to measure audio program loudness and true-peak audio level" ARIB technical data, ARIB TR-B32, "Loudness operation rules for digital television broadcast programs"

  As described above, in the conventional channel number conversion method, a constant value based on the reproduction position of the audio signal of each channel is used as a conversion coefficient for converting the number of channels regardless of the type of audio signal. For this reason, particularly when the viewer of the program is an elderly person, it may be difficult to hear the dialog due to the influence of the background sound in the audio signal after channel number conversion depending on the program.

  In addition, Recommendation ITU-R BS. In the loudness measurement method defined in 1770-3, the audio signal of each channel is multiplied by different weighting factors according to the reproduction position. Because the weighting coefficient according to the playback position and the coefficient used for channel number conversion are different, the average loudness value changes before and after the channel number is converted, and the average loudness value of all sections of the program is within the tolerance of the target loudness value. There were times when it did not fit.

  Therefore, the present invention has been made to solve the above-mentioned problem, and its object is to make the average loudness value constant before and after converting the number of channels, and to keep the ease of listening to the dialog. A channel number conversion device, a broadcast program reception device, and a program sound channel number conversion program.

  In order to achieve the above object, a program audio channel number converter according to the present invention converts the number of channels of an input audio signal of a program and generates an output audio signal. Pre-conversion calculation means for calculating the size of the sound of all channels, the size of the dialog contained in the input audio signal, and the size of the background sound contained in the input audio signal, and the number of channels of the input audio signal Means for converting the number of channels and generating an audio signal after conversion of the number of channels, and the size of sound of all channels in the audio signal after conversion of the number of channels generated by the number of channels conversion means; Size of the dialog included in the audio signal of the background sound, and the background sound included in the audio signal after the channel number conversion The post-conversion calculation means for calculating the magnitudes respectively, and the sizes of the sounds of all the channels match the sizes of the sounds of all the channels calculated by the pre-conversion calculation means, and the size of the dialog is the pre-conversion Adjusting means for adjusting the level of the audio signal after channel number conversion generated by the channel number conversion means so as to coincide with a predetermined value equal to or larger than the size of the dialog calculated by the calculation means It features.

  In the program audio channel number conversion device according to the present invention, the pre-conversion loudness value calculation means instead of the pre-conversion calculation means, the post-conversion loudness value calculation means instead of the post-conversion calculation means, and the adjustment means And the pre-conversion loudness value calculating means, for the input audio signal, the average loudness values of all channels and the average loudness of the dialog contained in the input audio signal for all time intervals of the program. Value and the average loudness value of the background sound included in the input audio signal are calculated respectively, and the converted loudness value calculating means calculates the number of channels after conversion of the audio signal after conversion by the channel number converting means. Average loudness value of all channels for all time segments of the program, after conversion of the number of channels The average loudness value of the dialog contained in the voice signal and the average loudness value of the background sound contained in the audio signal after the channel number conversion are respectively calculated, and the loudness adjustment means calculates the average loudness values of all the channels before the conversion. The average loudness value of the dialog matches the average loudness value of all the channels calculated by the loudness value calculation means, and the average loudness value of the dialog matches the predetermined loudness value of the dialog calculated by the pre-conversion loudness value calculation means And adjusting the level of the audio signal after channel number conversion generated by the channel number conversion means.

  The apparatus for converting the number of program audio channels according to the present invention further includes metadata extraction means, wherein the metadata extraction means extracts metadata from the input audio signal, and the average loudness value is extracted from the metadata. It is data used when calculating, and the channel determination information on the reproduction position of the audio signal of each channel before and after the channel number conversion, and the dialog determination value indicating whether the audio signal of each channel is the dialog or the background sound. It is characterized by extracting.

  The apparatus for converting number of program audio channels according to the present invention further includes channel separating means, and the channel separating means includes a dialog component and a background sound component among audio signals of each channel constituting the input audio signal. It is characterized in that the dialog which is the audio signal of the dialog component and the background sound which is the audio signal of the background sound component are separated from the audio signal of the channel.

  In the program audio channel number converter according to the present invention, at least one of the center channel, left channel, light channel, upper layer center channel and lower layer center channel among the channels of the input audio signal is a dialog channel And treating the audio signal of the dialog channel as the dialog included in the input audio signal.

  Further, the program audio channel number conversion device according to the present invention performs the above-mentioned assumption processing based on a preset designation, and when the designation indicates that the designation processing is not performed, the above-mentioned assumption processing is not performed. The pre-conversion loudness value calculating means calculates, for the input audio signal, only the loudness values of all channels for the entire time interval of the program, and the post-conversion loudness value calculating means generates the channel number converting means. Only the loudness value of all channels is calculated for the entire time interval of the program for the audio signal after the channel number conversion, and the loudness adjustment means calculates the average loudness value of all channels by the pre-conversion loudness value calculation means In the channel number conversion means to match the calculated average loudness value of all the channels Adjusting the level of the audio signal after the number of channels converted generated Ri, characterized in that.

  Furthermore, the program sound channel number conversion program according to the present invention causes a computer to function as the program sound channel number conversion device.

  A broadcast program receiving apparatus according to the present invention is a broadcast program receiving apparatus that receives a video signal and an audio signal of a broadcast program and reproduces the video signal and the audio signal, and includes the program audio channel number conversion apparatus. It is characterized by

  As described above, according to the present invention, it is possible to keep the average loudness value constant and to keep the dialog easy to hear before and after converting the number of channels.

It is a block diagram showing composition of a program sound channel number conversion device by an embodiment of the present invention. It is a flow chart which shows processing of a program sound channel number conversion device. It is a block diagram which shows the structural example of a pre-conversion loudness value calculation means. It is a figure explaining the loudness value adjusted by the loudness adjustment means. It is a block diagram showing an example of composition of loudness adjustment means. It is a flow chart which shows processing of loudness adjustment means. It is a figure explaining the loudness value and output sound signal adjusted by the loudness adjustment means. It is a block diagram which shows the structure of the program sound channel number conversion apparatus by other embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. According to the present invention, when converting the number of channels of the audio signal, the loudness of all the channels before and after the conversion of the number of channels match, and the size of the dialog before and after the conversion of the number of channels becomes the same, or It is characterized in that the level of the audio signal after channel number conversion is adjusted so that the size of the dialog after conversion is larger than that before channel number conversion.

[Program audio channel number converter]
First, a program sound channel number conversion apparatus according to an embodiment of the present invention will be described. FIG. 1 is a block diagram showing a configuration of a program sound channel number conversion device according to an embodiment of the present invention, and FIG. 2 is a flowchart showing processing of the program sound channel number conversion device. This program sound channel number conversion device 1 comprises: pre-conversion loudness value calculation means (pre-conversion calculation means) 11, channel number conversion means 12, post-conversion loudness value calculation means (post-conversion calculation means) 13 and loudness adjustment means (adjustment means) ) Has 14).

  In the program sound channel number conversion device 1, an input sound signal which is a program sound produced by N channels, a reproduction position of the sound signal of each channel before the channel number conversion, and a sound of each channel after the channel number conversion Channel arrangement information on the reproduction position of the signal, a dialog determination value indicating whether the audio signal of each channel is a dialog or background sound, and a target loudness value which is a target value of the average loudness value in all sections of the program It is input. Then, the program sound channel number conversion device 1 converts an audio signal of N channels into an audio signal of M channels, and outputs the program audio converted into M channels as an output audio signal. Here, M and N are generally different positive integers, but when the reproduction position is different, M and N may be the same.

  The channel allocation information is input to the program sound channel number conversion device 1 as preset information, and the dialog determination value and the target loudness value are input to the program sound channel number conversion device 1 as preset values.

(Pre-conversion loudness value calculation means 11)
The pre-conversion loudness value calculation means 11 inputs the input sound signal of the number N of channels, the channel arrangement information and the dialog determination value, and the audio channel of each channel constituting the input audio signal is a dialog channel based on the dialog determination value. The audio signal of each channel is classified by distributing it into the audio signal (dialog) of (1) and the audio signal (background sound) of the background sound channel (step S201).

  The pre-conversion loudness value calculating means 11 calculates the average loudness values of all channels for the entire channel of the audio signal of all channels using weighting factors based on channel allocation information (step S202). For the speech signal of the dialog channel classified in step S201, the average loudness value of the dialog is calculated for all sections of the program using a weighting factor or the like based on the channel arrangement information (step S203). Further, the pre-conversion loudness value calculating means 11 calculates the average loudness value of the background sound with respect to the sound signal of the background sound channel classified in step S201 using weighting factors based on the channel arrangement information for all sections of the program. Is calculated (step S204).

  By this, the average loudness value of all channels before channel number conversion, the average loudness value of dialog before channel number conversion, and the average loudness value of background sound before channel number conversion are calculated in the pre-conversion loudness value calculation means 11 And is output to the loudness adjustment means 14 as the loudness value before conversion.

  FIG. 3 is a block diagram showing a configuration example of the pre-conversion loudness value calculation means 11, and shows the case where the number of channels of the input sound signal is 5 (N = 5). The pre-conversion loudness value calculating means 11 comprises a filtering stage 21, a mean square stage 22, a weighting stage 23, a classification stage 24, an adding stage 25 and an absolute and relative gating stage 26.

  The filter processing stage 21 inputs the audio signal of each channel, and performs K characteristic filter processing on the audio signal for each channel. The root mean square stage 22 calculates the root mean square value of the audio signal subjected to the K characteristic filtering process by the filtering stage 21 for each channel.

  The weight processing stage 23 determines the weighting factor of each channel based on the reproduction position of the audio signal of each channel before the channel number conversion included in the channel layout information, and is calculated by the mean square stage 22 for each channel. The mean square value is multiplied by the weighting factor of the corresponding channel. Here, the weighting factor corresponding to the reproduction position for each channel is held in advance in the weight processing stage 23, and the weighting factor for each channel is determined according to the reproduction position of the audio signal of each channel and the weighting factor held in advance. It is determined. Thereby, the root mean square value multiplied by the weighting factor is obtained for each channel.

  The classification stage 24 receives the square average value multiplied by the weighting factor for each channel from the weight processing stage 23, and classifies the audio signal of each channel into a dialog or background sound based on the dialog determination value. Then, the classification stage 24 outputs the root mean square value of all channels, the root mean square value of the dialog and the root mean square value of the background sound to the adding stage 25.

  The addition stage 25 inputs from the classification stage 24 the root mean square value of all channels, the root mean square value of the dialog, and the root mean square value of the background sound. Then, the adding stage 25 adds the root mean square values of all channels to obtain an instantaneous root mean square value of all channels. Similarly, the addition stage 25 adds the root mean square value of the dialog to obtain an instantaneous root mean square value of the dialog alone, and adds the root mean square value of the background sound to find an instantaneous root mean square value of the background sound only .

The absolute and relative gating stage 26 applies absolute gating processing and relative gating processing to the instantaneous value of the root mean square of all channels obtained by the adding stage 25, and the root mean value of all channels for all sections of the program. Calculate Z _t . The absolute and relative gating stage 26, according to the following equation to convert the mean square value Z _t of all channels in average loudness value L _t of all channels (LKFS).
[Equation 1]
L _t = −0.691 + ₁₀ log ₁₀ Z _t (1)

The absolute and relative gating stage 26 performs absolute gating processing and relative gating processing on the instantaneous value of the root mean square of the dialog determined by the adding stage 25 and calculates the root mean square value Z _d of the dialog for all sections of the program. Ask for Also, the absolute and relative gating stage 26 transforms the dialog's root mean square value Z _d into the dialog's average loudness value L _{d according} to the following equation:
[Equation 2]
L _d = −0.691 + ₁₀ log ₁₀ Z _d (2)

The absolute and relative gating stage 26 obtains the root mean square value Z _b of the background sound for the entire section of the program, similarly to the processing for obtaining the root mean square value Z _d and the loudness value L _d of the dialog, and The root mean square value Z _b of the background sound is converted to the average loudness value L _b of the background sound.
[Equation 3]
L _b = −0.691 + ₁₀ log ₁₀ Z _b (3)

The example of FIG. 3 shows the case where the audio signal of channel number 3 is a dialog, and the audio signals of channel numbers 1, 2, 4, and 5 are background sounds. An adding stage 25 adds the root mean square values of channel numbers 1 to 5 to obtain an instantaneous root mean square value of all channels, and a root mean square value of channel number 3 is an instantaneous root mean square value of only dialog The root mean square values of channel numbers 1, 2, 4, and 5 are added to obtain an instantaneous root mean square value of only the background sound. The absolute and the relative gating stage 26, the audio signal of the channel numbers 1 to 5 mean square value Z _t and the average loudness value L _t of all channels are obtained, mean square value of the dialog for the audio signal of the channel number 3 Z _d and average loudness value L _d is determined, the audio signal of the channel numbers 1, 2, 4 mean square value Z _b and average loudness value L _b of the background noise is determined.

This makes it possible to know the mixing balance between the speech signal of the dialog before the channel number conversion and the speech signal of the background sound. Average loudness value (square mean value Z _t and average loudness value L _t ) of all channels before channel number conversion calculated by the pre-conversion loudness value calculation means 11, average loudness value (square average value) of dialog before channel number conversion Z _d and the average loudness value L _d ), and the average loudness value of the background sound before conversion of the channel number (root mean value Z _b and average loudness value L _b ) are output to the loudness adjustment means 14 as the loudness values before conversion. Be done.

  In addition, Recommendation ITU-R BS. According to the loudness measurement method specified in 1770-3, the average loudness value filters the audio signal of each channel, obtains a root mean square value, multiplies weighting factors, adds all channels, and It is determined by performing absolute gating processing and relative gating processing. The filtering stage 21, the mean square stage 22, the weighting stage 23, the adding stage 25 and the absolute and relative gating stage 26 of the pre-conversion loudness value calculating means 11 perform processing according to this loudness measurement method. For details of these processes, refer to Recommendation ITU-R BS. 1770-3 and the technical data ARIB TR-B32 of the above-mentioned Non-Patent Document 6.

(Channel number conversion means 12)
Referring back to FIG. 1 and FIG. 2, the channel number conversion means 12 receives the input sound signal of channel number N, channel arrangement information and dialog determination value, and converts the channel number before conversion based on the channel arrangement information and dialog determination value. The channel arrangement (arrangement of the audio signal of each channel constituting the input audio signal) and the channel arrangement after the channel number conversion (arrangement of the audio signal of each channel constituting the audio signal after channel number conversion) are specified. Then, the number-of-channels conversion means 12 converts the number of channels of the input audio signal according to the channel arrangement before and after the specified number-of-channels conversion, and generates the audio signal after the number of channels conversion. . That is, the channel number conversion means 12 converts the number of channels of the whole program using the conversion coefficient set in advance, and also uses only the audio signal of the dialog channel using the conversion coefficient set in advance according to the dialog determination value. Convert the number of channels, and also convert the number of channels only for the audio signal of the background sound channel.

  Specifically, the number-of-channels conversion unit 12 converts the input audio signal of each channel into an audio signal of the dialog channel and an audio signal of the background sound channel based on the dialog determination value, as in the process of step S201. By distributing, the audio signal of each channel is classified. The number-of-channels conversion unit 12 arranges the channels for the audio signal of the dialog channel before and after the channel number conversion based on the classification of the audio signal in each channel and the channel arrangement information, and the sound of the background sound channel before and after the channel number conversion. Identify the channel arrangement for the signal.

For example, when converting the number of channels 5 into the number of channels 2, the channel number conversion means 12 converts conversion coefficients G _C , G _L , G _R preset according to the following equation according to the channel arrangement before and after the specified channel number conversion. The number of channels ₅ of the input audio signals C ₅ , L ₅ , R ₅ , Ls ₅ and Rs ₅ of all the input channels is converted into the number of channels 2 using G, G _LS and G _RS, and all the channels after channel number conversion The voice signals L ₂ and R _{2 of the above} are generated (step S205).
[Equation 4]
L ₂ = G _C C ₅ + G _L L ₅ + G _LS Ls ₅
R ₂ = G _C C ₅ + G _R R ₅ + G _RS Rs ₅ (4)
Here, the audio signal C ₅ is a voice signal of a dialog channel audio signals _{L 5, R 5, Ls 5} , Rs 5 is an audio signal of the channel background sound. As conversion coefficients in this case, G _C , G _LS , and G _RS = 0.7071, and G _L and G _R = 1.000 are used.

Number of channel conversion means 12, according to the channel arrangement of the audio signal of the dialog channels before and after the number of channel conversion identified by the following equation, using the conversion factor G _C set in advance, the voice signal of a dialog channel input C The channel number 1 of ₅ is converted to the channel number 2, and audio signals (audio signals of dialog) L _d2 and R _d2 of the dialog channel after conversion of the channel number are generated (step S206).
[Equation 5]
L _d2 = G _C C ₅
R _d2 = G _C C ₅ (5)

  Here, although the number of channels of the entire program is converted from 5 to 2, each audio signal of the number of channels 2 of the entire program after the number of channels conversion includes a dialog. Therefore, focusing on the dialog alone, the number of channels of the dialog channel before the channel number conversion is one, and the number of dialog channels after the channel number conversion is two.

The channel number conversion means 12 uses conversion coefficients G _LS , G _RS , G _L and G _R set in advance according to the following equation according to the channel arrangement for the background sound channel before and after the specified channel number conversion. The audio signal of the background sound channel L ₅ , R ₅ , Ls ₅ , Rs ₅ is converted to the channel number 2 and the audio signal of the background sound channel after the channel number conversion (audio signal of the background sound) ) L _b2 and R _b2 are generated (step S207).
[Equation 6]
L _b2 = G _L L ₅ + G _LS Ls ₅
R _b2 = G _R R ₅ + G _RS Rs ₅ (6)

  Here, the number of channels of the entire program is converted from 5 to 2, but background sound is included in each of the audio signals of the number 2 of channels of the whole program after the conversion of the number of channels. Therefore, focusing only on the background sound, the number of background sound channels before the channel number conversion is four, and the number of background sound channels after the number channel conversion is two.

  As a result, the channel number conversion means 12 generates audio signals of all channels after channel number conversion, audio signals of dialog channel after channel number conversion, and audio signals of background sound channel after channel number conversion. Is output to the loudness adjustment means 14 as an audio signal of

  The present invention does not limit the channel number conversion processing by the channel number conversion means 12 to the above-described processing. For example, as the conversion factor, the channel number conversion means 12 uses the recommendation ITU-R BS. A downmix coefficient defined in 775-3 or ARIB STD-B32 standard of Non-Patent Document 1 may be used. Also, the channel number conversion means 12 may obtain conversion coefficients in accordance with the angle depending on the channel position, and may convert the number of channels using this conversion coefficient.

(Converted loudness value calculation means 13)
The post-conversion loudness value calculation means 13 receives the channel arrangement information and also converts the audio signals after conversion from the channel number conversion means 12 (audio signals of all channels after channel number conversion, audio signals of dialog channel after channel number conversion And the audio signal of the background sound channel after channel number conversion).

  The post-conversion loudness value calculation means 13 calculates the average loudness values of all channels for all sections of the program for the audio signals of all channels after the channel number conversion corresponding to the channel arrangement information (step S208). For the speech signal of the dialog channel after number conversion, the average loudness value of the dialog is calculated for all sections of the program (step S209). Further, the converted loudness value calculating means 13 calculates the average loudness value of the background sound for the entire section of the program for the sound signal of the background sound channel after the channel number conversion (step S210).

The processes of steps S208 to S210 are the same as the processes of steps S202 to S204 and the process shown in FIG. 3, and for all sections of the program after channel number conversion, the root mean square value Z _{t2 of} all channels and An average loudness value L _t2 , a root mean square value Z _d2 and an average loudness value L _{d2 of} the dialog, and a root mean square value Z _b2 and an average loudness value L _{b2 of the} background sound are calculated.

Thereby, in the converted loudness value calculating means 13, the average loudness value (square mean value _Zt2 and average loudness value _Lt2 ) of all channels after channel number conversion, average loudness value of dialog after channel number conversion (square average) The value Z _d2 and the average loudness value L _d2 ), and the average loudness value of the background sound after conversion of the number of channels (the root mean square value Z _b2 and the average loudness value L _b2 ) are calculated, and the loudness adjustment means It is output to 14.

(Loudness adjustment means 14)
The loudness adjustment means 14 receives the target loudness value and converts the audio signal after conversion from the channel number conversion means 12 (audio signal of all channels after channel number conversion, audio signal of dialog channel after channel number conversion, and Input the audio signal of the background sound channel after channel number conversion. Further, the loudness adjustment means 14 receives the pre-conversion loudness value (average loudness value of all channels before conversion of channel number (square mean value Z _t and average loudness value L _t ) before conversion) from the pre-conversion loudness value calculation means 11, channel number conversion Input the average loudness value (root mean value Z _d and average loudness value L _d ) of the previous dialog, and the average loudness value (root mean value Z _b and average loudness value L _b ) of the background sound before channel number conversion And the converted loudness value from the converted loudness value calculating means 13 (average loudness value of all channels after conversion of the number of channels (square mean value _Zt2 and average loudness value _Lt2 ), average of the dialog after the number of channels conversion loudness value (root mean square value Z _d2 and average loudness value L _d2), and the average of the background noise after the number of channels converted Entering loudness value (root mean square value Z _b2 and average loudness value L _b2)).

  Here, when the loudness adjustment means 14 inputs the target loudness value and the converted audio signal and performs loudness adjustment, the pre-conversion loudness value calculation means 11 calculates the pre-conversion pre-calculated for all sections of the program. The loudness value is input, and the converted loudness value calculated in advance for all sections of the program is input from the converted loudness value calculating means 13 in advance.

  The loudness adjustment means 14 calculates target loudness values of the dialog and background sound based on the pre-conversion loudness value and the post-conversion loudness value (step S211), and adjusts the post-conversion audio signal (step S212). Generate an output audio signal. Specifically, the loudness adjustment means 14 matches the average loudness value with the target loudness value (average loudness value of all channels before channel number conversion) for the audio signals of all channels after channel number conversion, and the channels Adjust the level of the audio signal after channel number conversion so that the average loudness value of the dialog channel audio signal after number conversion matches the number before channel number conversion or is larger than that before channel number conversion, An audio signal with M channels after loudness adjustment is generated as an output audio signal.

FIG. 4 is a diagram for explaining the loudness value adjusted by the loudness adjusting means 14, and shows an example of converting the number of channels from 5 to 2. In FIG. Hereinafter, for convenience of explanation, the mean square value Z will be described as the loudness value. FIG. 4 (1) shows the loudness value Z _t representing the added value in the audio signal of all channels before the channel number conversion, the loudness value Z _{d of} dialog C representing the added value in the audio signal of the dialog channel, and the background sound channel background sound representative of the added value in the audio signal of the L, R, Ls, shows a loudness value Z _b of Rs.

FIG. 4 (2) shows the loudness value after converting the audio signal of channel 5 shown in FIG. 4 (1) into an audio signal of channel 2, and the audio signals of all channels after channel number conversion loudness value represents the sum value of Z _t2, loudness value Z _d2 dialog L _d2, R _d2 representing the sum value of the audio signal of the dialog channels, and background sound L _b2 representing the sum value of the audio signal of the background sound channel, It shows a loudness value Z _b2 of R _b2.

4 (1) as shown in (2), by converting the number of channels, loudness value Z _d of the dialog, Z _d2 does not change, loudness value Z _b2 background sound before conversion loudness value Z _b Assume that it becomes larger than In this case, as shown in FIG. 4 (3), so that the loudness value Z _t of all channels before loudness value Z _t2 and the number of channel conversion of all the channels after the number of channels converted match, after the number of channel conversion When simply adjusted while maintaining the component ratio of the audio signal, a loudness value Z _d2 dialog after the number of channels conversion becomes smaller than the loudness value Z _d of the previous number of channels conversion dialog.

Therefore, in the embodiment of the present invention, loudness adjustment means 14, a loudness value Z _t of all channels before loudness values for all channels after the number of channels converted Z _t2 and the number of channel conversion match and, FIG. 4 ( as shown in 4), so that the loudness value before loudness values dialog after the number of channels converted Z _d2 and the number of channel conversion Z _d are matched, to adjust the level of the audio signal after the number of channel conversion. This makes it possible to convert the number of channels without breaking the mixing balance between the dialog and the background sound.

4 (4) coincides with loudness value Z _t of all channels before loudness value Z _t2 and the number of channel conversion of all channels after channel number conversion and loudness value dialog after the number of channels converted Z _d2 And the loudness value Z _t2 representing the added value in the audio signals of all channels after the channel number conversion adjusted to match the loudness value Z _d of the dialog before the channel number conversion, and the added value in the audio signal of the dialog channel It shows Z _d2 and Z _b2 representing an addition value in the audio signal of the dialog channel.

In the embodiments of the present invention, loudness adjustment means 14, a loudness value Z _t of all channels before loudness values for all channels after the number of channels converted Z _t2 and the number of channel conversion match and, FIG. 4 ( as shown in 5), loudness value Z _d2 dialog after the number of channels conversion so that greater than loudness value Z _d before the number of channels conversion to adjust the level of the audio signal after the number of channel conversion.

4 (5) coincides with loudness value Z _t of all channels before loudness value Z _t2 and the number of channel conversion of all channels after channel number conversion and loudness value dialog after the number of channels converted Z _d2 The loudness value Z _t2 representing the addition value in the audio signals of all channels after adjustment of the number of channels after adjustment so as to be larger than the average loudness value Z _d of the dialog before the channel number conversion The loudness value Z _{d2 is shown} , and the loudness value Z _b2 is shown, which represents the addition value in the audio signal of the background sound channel.

As in this example, by reducing the number of channels from 5 to 2, the dialog and the background sound are converted to the same channel and reproduced from the same speaker. In this case, since the sound source positions of the dialog and the background sound overlap spatially and space masking occurs, the dialog may be difficult to hear. Therefore, to increase the loudness value Z _d2 dialog after the number of channels converted, by decreasing the loudness value Z _b2 background sound after the number of channels conversion, it is possible to solve such problems.

  FIG. 5 is a block diagram showing a configuration example of the loudness adjustment means 14, and FIG. 6 is a flowchart showing processing of the loudness adjustment means 14. As shown in FIG. FIG. 6 shows an example of converting the number of channels from 5 to 2 as in FIG.

  The loudness adjusting means 14 includes a target loudness value calculating stage 31 and a level adjusting stage 32. The loudness adjustment means 14 receives the converted audio signal from the channel number conversion means 12, the loudness value before and after conversion, and the target loudness value from the pre-conversion loudness value calculation means 11 and the post-conversion loudness value calculation means 13 (step S601). ).

Specifically, the target loudness value calculation stage 31, a pre-conversion loudness value calculating means 11, the loudness value of all channels before the channel number converting Z _t, a loudness value Z _b loudness value Z _d and background sound dialog and inputs, from the converted loudness value calculating means 13, a loudness value Z _t2 of all channels after channel number conversion, enter the loudness value Z _b2 loudness value Z _d2 and background sound of the dialog, the target loudness value Z _tt input. Further, the level adjustment stage 32 receives the audio signals L ₂ , R ₂ , L _d2 , R _d2 , L _b2 , R _b2 (L ₂ = L _d2 + L _b2 , R ₂ =) after the channel number conversion means 12 converts the channel number. Enter R _d2 + R _b2 ).

Here, loudness value Z _t of all channels before the channel number conversion is the same as the target loudness value Z _tt. Further, the audio signals L ₂ and R ₂ are audio signals of all channels after the channel number conversion as in the equation (4), and the audio signals L _d2 and R _d2 are channels as in the equation (5) The voice signal of the dialog channel after number conversion is obtained, and the voice signals L _b2 and R _b2 are voice signals of the background sound channel after channel number conversion, as described in the equation (6).

Target loudness value calculation stage 31 sets the target loudness value Z _tt as the target loudness value Z _t2A of all channels after loudness adjustment (step _{S602: Z t2A = Z tt (} = Z t)). Then, the target loudness value calculation stage 31, as the target loudness value Z _d2A dialog after loudness adjustment, equal to the mean square value Z _d of the previous number of channels converted dialog or mean square value greater predetermined value than Z _{d, Are} set (step S603: Z _d2A ZZ _d ). This predetermined value is larger than the root mean square value Z _d of the dialog before the channel number conversion and smaller than the target loudness value Z _tt (= Z _t ).

The target loudness value calculation stage 31 subtracts the target loudness value _Zd2A of the dialog set in step S603 from the target loudness values _Zt2A of all channels set in step S602, and the subtraction result is the background sound after the loudness adjustment. set of the target loudness value Z _B2a (step _{S604: Z b2A = Z t2A -Z} d2A).

Target loudness value calculation stage 31, the target loudness value Z _t2A of all channels set in step S602, the dialog of the target loudness value Z _d2A set in step S603, and the target loudness value of the background sound set in step S604 _Output Z _b2A to the level adjustment stage 32.

The level adjustment stage 32 receives the target loudness values _Zt2A of all channels, the dialog target loudness value _Zd2A and the background sound target loudness value _Zb2A from the target loudness value calculation stage 31. The level adjustment stage 32, to the audio signal after the conversion input from the channel number converting unit 12, a loudness value Z _t2 of all channels of the audio signal is equal to the target loudness value Z _t2A of all channels, and, as loudness value Z _d2 dialog of the audio signal coincides with the target loudness value Z _d2A dialog, it adjusts the level of the audio signal (step S605).

FIG. 7 is a diagram for explaining the loudness value and the output sound signal adjusted by the loudness adjustment means 14. As shown on the left side of FIG. 7, according to the process of step S605 shown in FIG. 6, an audio signal corresponding to the loudness value after the loudness adjustment, that is, the loudness value _{Zt2A of} all channels (FIG. 4 (4) (5) resulting in Z _t2), the dialog loudness value Z _d2A (FIG. 4 (4) audio signal (5) in Z _d2) and the background sound loudness value Z _B2a (FIG. 4 (4) (5) in the Z _b2) is Be This audio signal is _composed of dialogs L _d2A and R _d2A corresponding to the loudness value Z _d2A of the dialog and background sounds L _b2A and R _b2A corresponding to the loudness value Z _b2A of the background sound.

Returning to FIG. 6, as shown on the right side of FIG. 7, the level adjustment stage 32 generates an audio signal L _{2 A} composed of the dialog L _{d 2 A} and the background sound L _{b 2 A} adjusted in step S 605. by adjusting the dialog R _d2A and background sound R _B2a at S605 to generate a composed audio signal R _2A. The level adjustment stage 32 outputs the audio signals L _2A of the channel number 2, the R _2A as the output audio signal is an audio signal after loudness adjustment (step S606). The addition process for generating the audio signals L _2A and R _2A corresponds to the equation (4).

Thus, the loudness adjustment means 14, loudness value of all channels equal to the target loudness value Z _t2A of all channels (loudness value of all channels before the channel number converting Z _t), and, loudness value dialog dialog Dialog L _d2A , R _{d2 A} and background sound L _{b2 A} , R that constitute the audio signal after conversion so as to match the target loudness value Z _d2A (a predetermined value of the loudness value Z _d of the dialog before channel number conversion) to adjust the level of _B2a, audio signal of the channel number _{2 L 2A (= L d2A +} L b2A), and outputs the _{R 2A (= R d2A + R} b2A) as an output audio signal is an audio signal after loudness adjustment.

The level adjustment stage 32 of the loudness adjustment means 14, in step S605 of FIG. 6, with respect to the audio signal after the conversion, loudness value Z _t2 of all channels of the audio signal and the target loudness value Z _t2A of all channels one We, and as loudness value Z _d2 dialog of the audio signal coincides with the target loudness value Z _d2A dialog, and to adjust the level of the audio signal. In contrast, the level adjustment stage 32, to the audio signal after the conversion, loudness value Z _d2 dialog of the audio signal coincides with the target loudness value Z _d2A dialog (in the example of FIG. 4, after the number of channel conversion Dialog loudness value Z _d2 matches the loudness value Z _d of the dialog before channel number conversion, or a value greater than the loudness value Z _d ), and the loudness value Z _b2 of the background sound of the audio signal is the target of the background sound in the example shown in (Fig. 4 to match the loudness value Z _B2a, loudness value Z _b2 background sound after the number of channels converted is smaller than the loudness value Z _b or loudness value Z _b, the background sound of the front channel number conversion The level of the audio signal may be adjusted to match the value).

In short, the loudness level adjustment stage 32 of the adjusting means 14, loudness value of all channels to match the loudness value Z _t of all channels before the channel number conversion, and loudness loudness value before the number of channels converted Dialog Dialog so as not to be smaller than the value Z _d, may be adjusted level of the audio signal after the conversion.

  As described above, according to the program sound channel number conversion device 1 of the embodiment of the present invention, the pre-conversion loudness value calculation means 11 generates dialog and background for the input sound signal with N channels based on the dialog determination value. The average loudness value of all channels, the average loudness value of the dialog, and the average loudness value of the background sound are calculated using the weighting factors based on the channel arrangement information, etc., classified into the sound and for all sections of the program. .

  The channel number conversion means 12 specifies the channel arrangement before the channel number conversion and the channel arrangement after the channel number conversion based on the channel arrangement information, and conversion coefficients set in advance according to the specified channel arrangement before and after the channel number conversion. To convert the number N of channels of the input audio signal into the number M of channels of the output audio signal, and generate the audio signals of all channels after conversion of the number of channels, the audio signal of the dialog channel and the audio signal of the background sound channel I made it. The post-conversion loudness value calculation means 13 calculates the average loudness values of all channels for all sections of the program with respect to the audio signals of all channels after conversion of the number of channels, the audio signals of the dialog channel and the audio signals of the background sound channel. The average loudness value of the dialog and the average loudness value of the background sound are calculated.

  The loudness adjustment means 14 determines that the average loudness value of all channels matches the average loudness value of all channels before the channel number conversion, and the average loudness value of the dialog is a predetermined value greater than the average loudness value of the dialog before channel number conversion. To adjust the levels of the dialog and background sound that make up the audio signal after channel number conversion, and output the audio signal with M channels as the output audio signal that is the audio signal after loudness adjustment .

  As a result, since the average loudness value of the audio signal before and after channel number conversion can be made constant, the average loudness value of all sections of the program is within the allowable range of the target loudness value for the audio signal before channel number conversion. If so, the average loudness value of the entire section of the program can be within the tolerance of the target loudness value even for the audio signal after channel number conversion. Further, since the average loudness value of the dialog does not become smaller than the average loudness value of the dialog before the channel number conversion, it becomes possible to maintain the ease of hearing the dialog about the audio signal before and after the channel number conversion.

[Program audio channel number converter / other embodiment]
Next, a program sound channel number conversion apparatus according to another embodiment of the present invention will be described. FIG. 8 is a block diagram showing the configuration of a program sound channel number conversion apparatus according to another embodiment of the present invention. The program sound channel number conversion device 2 includes pre-conversion loudness value calculation means 11, channel number conversion means 12, post-conversion loudness value calculation means 13, loudness adjustment means 14, metadata extraction means 15, and channel separation means 16. There is. The program audio channel number conversion device 2 separates the dialog and the background sound from the audio signal of the channel including the dialog component and the background sound component among the input audio signal of the channel number N, and the channel from the metadata included in the input audio signal The arrangement information, the dialog determination value and the target loudness value are extracted.

  Comparing the program audio channel number converter 1 according to the embodiment of the present invention shown in FIG. 1 with the program audio channel number converter 2 according to the other embodiment of the present invention shown in FIG. , 2 are identical in that the pre-conversion loudness value calculating means 11, the channel number converting means 12, the post-conversion loudness value calculating means 13 and the loudness adjusting means 14 are provided. In addition to the configuration of the program sound channel number conversion device 1, a metadata extraction unit 15 and a channel separation unit 16 are further included. Also, the input audio signal of the number N of channels in the program audio channel number conversion device 1 shown in FIG. 1 is constituted by the audio signal of the dialog dedicated channel and the audio signal of the background sound dedicated channel. The difference is that the input audio signal of the channel number N in the audio channel number conversion device 2 includes an audio signal of a channel including the dialog component and the background sound component. In the program audio channel number converter 2 shown in FIG. 8, the parts common to the configuration of the program audio channel number converter 1 shown in FIG.

  The metadata extraction unit 15 receives an input audio signal of the number N of channels, extracts metadata from the input audio signal, and extracts channel arrangement information, a dialog determination value, and a target loudness value from the metadata. Then, the metadata extraction unit 15 outputs the extracted channel arrangement information to the pre-conversion loudness value calculation unit 11, the channel number conversion unit 12, and the post-conversion loudness value calculation unit 13, and the extracted dialog determination value is the pre-conversion loudness value. The target loudness value extracted is output to the loudness adjustment unit 14, and output to the calculation unit 11, the channel number conversion unit 12, and the channel separation unit 16. The channel arrangement information and the dialog determination value are used when the average loudness value is calculated in the pre-conversion loudness value calculation means 11 and the post-conversion loudness value calculation means 13. Also, the dialog determination value is used when separating channels in the channel separating means 16 described later. Furthermore, the target loudness value is used in the loudness adjustment in the loudness adjustment means 14.

  Here, the dialog determination value includes, in addition to the value indicating whether the audio signal of each channel is the dialog or the background sound, a value indicating that the audio signal includes the dialog component and the background sound component. I assume.

  The channel separation means 16 receives an input audio signal of the number N of channels and also receives a dialog determination value from the metadata extraction means 15, and based on the dialog determination value, an audio signal of a channel including a dialog component and a background sound component. Identify Then, the channel separating means 16 separates the audio signal of the dialog component and the audio signal of the background sound component from the audio signal, and the pre-conversion loudness value calculating means 11 as an audio signal of the dialog channel and an audio signal of the background sound channel. Output to This separation method (a method of separating an audio signal of a dialog component and an audio signal of a background sound component from an audio signal of a channel including a dialog component and a background sound component) is known, and there are various methods. The channel separation means 16 may use any separation method, regardless of the method.

  Further, the channel separation unit 16 specifies the audio signal of the dialog channel and the audio signal of the background sound channel based on the dialog determination value, and outputs the audio signal of the specified channel to the pre-conversion loudness value calculation unit 11 as it is.

  As described above, according to the program sound channel number conversion device 2 of the other embodiment of the present invention, in addition to the configuration of the program sound channel number conversion device 1, the metadata extraction unit 15 The channel arrangement information, the dialog determination value and the target loudness value are extracted from the data, and the channel separating means 16 separates the dialog and the background sound from the audio signal of the channel including the dialog component and the background sound component.

  As a result, even if an audio signal of a channel including a dialog component and a background sound component is present in the input audio signal, the channel arrangement information, the dialog determination value and the target loudness value are not set in advance. However, the pre-conversion loudness value calculation means 11 can calculate the pre-conversion loudness value, and the channel number conversion means 12 can convert the number of channels of the input audio signal, and the post-conversion loudness value calculation means 13 can calculate the converted loudness value. Therefore, as in the program sound channel number conversion device 1 according to the embodiment of the present invention, the average loudness value can be made constant before and after the conversion of the channel number, and the easiness of hearing the dialog can be maintained.

  The hardware configuration of the program sound channel number conversion device 1 according to the embodiment of the present invention and the hardware configuration of the program sound channel number conversion device 2 according to another embodiment of the present invention may use a normal computer. it can. The program sound channel number conversion devices 1 and 2 are configured by a computer including a CPU, a volatile storage medium such as a RAM, a non-volatile storage medium such as a ROM, and an interface. Each function of the pre-conversion loudness value calculation means 11, the channel number conversion means 12, the post-conversion loudness value calculation means 13 and the loudness adjustment means 14 provided in the program sound channel number conversion device 1 Each is realized by executing. The pre-conversion loudness value calculation means 11, the channel number conversion means 12, the post-conversion loudness value calculation means 13, the loudness adjustment means 14, the metadata extraction means 15 and the channel separation means 16 included in the program sound channel number conversion device 2. Each function is realized by causing a CPU to execute a program in which these functions are described. These programs are stored in the storage medium and read out and executed by the CPU. These programs can also be stored and distributed in storage media such as magnetic disks (floppy (registered trademark) disks, hard disks, etc.), optical disks (CD-ROM, DVD, etc.), semiconductor memories, etc. Can also be sent and received.

  Although the present invention has been described above by the embodiment and the other embodiments, the present invention is not limited to the embodiments and the other embodiments, and various modifications can be made without departing from the technical concept thereof. It is. For example, in the embodiment and the other embodiments, the pre-conversion loudness value calculation means 11 and the post-conversion loudness value calculation means 13 of the program sound channel number conversion devices 1 and 2 are the same as those described in Recommendation ITU-R of Non-Patent Document 5. BS. The average loudness value is calculated by the loudness measurement method specified in 1770-3, but instead of the average loudness value, the loudness such as the average sound pressure level, 5% time rate average sound pressure level, etc. An index representing 算出 may be calculated. The average loudness value is a numerical value corresponding to the subjective quantity of the loudness, whereas the average sound pressure level is a numerical value corresponding to the objective quantity of the loudness.

  In this case, the pre-conversion loudness value calculating unit 11 calculates the sizes of the sounds of all channels in the input audio signal, the size of the dialog included in the input audio signal, and the size of the background sound included in the input audio signal. The post-conversion loudness value calculating means 13 calculates the sizes of the sounds of all channels in the converted audio signal whose number of channels has been converted by the number-of-channels converting means 12, the size of the dialog included in the converted audio signal, And the size of the background sound included in the converted audio signal. Then, the loudness adjustment means 14 calculates the level of the audio signal after the channel number conversion based on the loudness before and after the channel number conversion calculated by the pre-conversion loudness value calculation means 11 and the post-conversion loudness value calculation means 13. adjust.

  Further, in the above-described embodiment and the other embodiments, the pre-conversion loudness value calculation means 11 and the post-conversion loudness value calculation means 13 of the program sound channel number conversion devices 1 and 2 calculate average loudness values for all sections of the program. Although the calculation is made, the short time loudness value for each constant time may be calculated. The loudness adjustment means 14 is an audio signal after conversion of the number of channels at each constant time based on the short time loudness value for each constant time calculated by the pre-conversion loudness value calculation means 11 and the post-conversion loudness value calculation means 13 Adjust the level of This makes it possible to perform sequential processing corresponding to live broadcast programs and the like.

  Further, in the other embodiment, the program sound channel number conversion device 2 includes the pre-conversion loudness value calculation unit 11, the channel number conversion unit 12, the post-conversion loudness value calculation unit 13, the loudness adjustment unit 14, and the metadata extraction unit 15 And although the channel separating means 16 is provided and configured, the metadata extracting means 15 may be removed from these components. The channel arrangement information, the dialog determination value, and the target loudness value are preset, as in the program sound channel number conversion device 1 shown in FIG. As a result, the program audio channel number conversion device 2 does not need to extract metadata from the input audio signal, and does not need to extract channel arrangement information, dialog determination value and target loudness value from metadata, so processing load It can be reduced.

  Furthermore, the program audio channel number conversion device 2 may be configured by removing the channel separation means 16 from the components shown in FIG. The audio signal of each channel in the input audio signal is either a dialog or a background sound.

  In the above embodiment, the case where the input audio signal includes the channel of the dialog is described, and in the other embodiment, the dialog component and the background sound component are included in the audio signal of at least one channel constituting the input audio signal. The case of inclusion is described. These assume that the dialog can be distinguished from the input speech signal. However, in the above embodiment, when the user can not obtain channel arrangement information etc. in advance, or in the other embodiment, there may be a case where metadata is not added to the input voice signal. Moreover, even if the sound source separation technique by the channel separation means 16 of the program sound channel number conversion device 2 is used in the other embodiment, the dialog may not be separated from the input sound signal. In this case, the dialog can not be distinguished from the input speech signal.

  Therefore, the pre-conversion loudness value calculation means 11 and the number-of-channels conversion means 12 of the program sound channel number conversion devices 1 and 2 include the center channel (C-ch: Centre channel), center channel and channels on both sides of the input sound signal. L-ch (left channel): Left channel, R-ch (right channel): Right channel, or center channel and upper / lower center channel (TpFC: Top front center channel / Cv: center front vertical height, BtFC : Bottom front center channel / Cb: Center front vertical bottom) It may be performed regarded processing treated as a channel. Also, the pre-conversion loudness value calculation means 11 and the channel number conversion means 12 treat as a dialogue channel at least one of the center channel, left channel, light channel, upper layer center channel and lower layer center channel May be performed. Generally, the audio signals of these channels contain many dialog components.

  In this case, the pre-conversion loudness value calculation means 11 and the number-of-channels conversion means 12 of the program sound channel number conversion devices 1 and 2 may each be provided with a switching stage. The switching stage receives the input audio signal, and when the switching specification set in advance is the setting for setting the dialog channel, the above-mentioned processing for only the dialog channel is performed. On the other hand, when the switching specification set in advance is the specification of not setting the dialog channel, the switching stage does not perform the above-mentioned processing only for the dialog channel. The pre-conversion loudness value calculation means 11 calculates only the loudness values of all the channels before the channel number conversion, and the post-conversion loudness value calculation means 13 calculates only the loudness values of all the channels after the channel number conversion. Also, the loudness adjustment means 14 adjusts the level of the audio signal after the channel number conversion so that the loudness values of all the channels match the loudness values of all the channels before the channel number conversion.

  Also, even when audio signals of all channels in the input audio signal do not contain dialog components, the pre-conversion loudness value calculation means 11 of the program audio channel number conversion devices 1 and 2 calculates the loudness of all channels before channel number conversion. Only the values are calculated, and the converted loudness value calculating means 13 calculates only the loudness values of all the channels after the channel number conversion, and the loudness adjusting means 14 calculates the loudness values of all the channels of all the channels before the channel number conversion. Adjust the level of the audio signal after channel number conversion to match the loudness value. The pre-conversion loudness value calculation means 11, the post-conversion loudness value calculation means 13 and the loudness adjustment means 14 determine that the audio signals of all channels in the input audio signal do not include a dialog component based on the dialog determination value.

  Furthermore, the program audio channel number conversion devices 1 and 2 can be applied to a broadcast program reception device that receives video signals and audio signals of broadcast programs and reproduces them. The broadcast program receiving apparatus is configured to include the program audio channel number converter 1 shown in FIG. 1 or the program audio channel number converter 2 shown in FIG.

  The program audio channel number converter according to the present invention is useful in audio format conversion at the time of program production or program exchange, or adaptive conversion to a reproduction system in a broadcast program receiver.

1, 2 Program audio channel number converter 11 Pre-conversion loudness value calculation means (pre-conversion calculation means)
12 channel number conversion means 13 post-conversion loudness value calculation means (post-conversion calculation means)
14 loudness adjustment means (adjustment means)
15 metadata extraction means 16 channel separation means 21 filter stage 22 square average stage 23 weight process stage 24 classification stage 25 addition stage 26 absolute and relative gating stage 31 target loudness value calculation stage 32 level adjustment stage

Claims (8)

In a program audio channel number conversion device for converting the number of channels of an input audio signal of a program and generating an output audio signal,
Pre-conversion calculation means for calculating the sizes of the sound of all channels in the input audio signal, the size of the dialog included in the input audio signal, and the size of the background sound included in the input audio signal;
Number-of-channels conversion means for converting the number of channels of the input audio signal and generating the audio signal after the number of channels conversion;
The size of the sound of all channels in the audio signal after channel number conversion generated by the channel number conversion means, the size of the dialog included in the audio signal after the channel number conversion, and the audio signal after the channel number conversion After-conversion calculation means for calculating the size of the background sound included in
The loudness of all channels matches the loudness of all channels calculated by the pre-conversion calculation means, and the dialog size is greater than the dialog size calculated by the pre-conversion calculation means An adjusting means for adjusting the level of the audio signal after channel number conversion generated by the channel number converting means so as to match a predetermined value;
A program audio channel number conversion device characterized by comprising. In the program audio channel number conversion device according to claim 1,
A pre-conversion loudness value calculation means is provided instead of the pre-conversion calculation means, a post-conversion loudness value calculation means is provided instead of the post-conversion calculation means, and a loudness adjustment means is provided instead of the adjustment means.
The pre-conversion loudness value calculation means
The average loudness value of all channels, the average loudness value of the dialog included in the input audio signal, and the average loudness value of the background sound included in the input audio signal for the input audio signal over the entire time interval of the program Calculate each,
The post-conversion loudness value calculation means
With respect to the audio signal after channel number conversion generated by the channel number conversion means, average loudness value of all channels and average loudness value of dialog contained in the audio signal after channel number conversion over the entire time interval of the program And calculating the average loudness value of the background sound included in the audio signal after the channel number conversion,
The loudness adjustment means
The average loudness values of all the channels correspond to the average loudness values of all the channels calculated by the pre-conversion loudness value calculation means, and the average loudness values of the dialog are the averages of the dialogs calculated by the pre-conversion loudness value calculation means And converting the level of the audio signal after channel number conversion generated by the channel number conversion means so as to coincide with a predetermined value equal to or greater than a loudness value. In the program audio channel number conversion device according to claim 2,
Furthermore, metadata extraction means are provided,
The metadata extraction unit
Data used for extracting metadata from the input audio signal and calculating the average loudness value from the metadata, the channel arrangement information on the reproduction position of the audio signal of each channel before and after the conversion of the number of channels, An apparatus for converting the number of program audio channels, which extracts a dialog determination value indicating whether the audio signal of each channel is a dialog or a background sound. In the program audio channel number conversion device according to claim 2 or 3,
Furthermore, a channel separating means is provided,
The channel separation means is
From the audio signal of the channel including the dialog component and the background sound component among the audio signals of each channel constituting the input audio signal, the dialog which is the audio signal of the dialog component and the background sound which is the audio signal of the background sound component And an apparatus for converting the number of program audio channels. In the program sound channel number conversion device according to any one of claims 2 to 4,
An audio signal of the dialog channel is processed by treating at least one of the center channel, left channel, right channel, upper center channel and lower center channel among the channels of the input audio signal as the dialog channel. The program audio channel number converter, wherein the dialog is included in the input audio signal. In the program audio channel number conversion device according to claim 5,
Perform the above-mentioned ignoring process based on a preset designation,
If the designation indicates that no categorization processing is performed, the pre-conversion loudness value calculating means may calculate the loudness of all channels for all the time sections of the program with respect to the input audio signal without performing the categorization processing. Only the value is calculated, and the converted loudness value calculating means calculates only the loudness values of all the channels for the entire time interval of the program with respect to the audio signal after the channel number conversion generated by the channel number converting means. The loudness adjustment means converts the number of channels converted by the channel number conversion means so that the average loudness values of all channels coincide with the average loudness values of all channels calculated by the pre-conversion loudness value calculation means. Adjust the audio signal level of the program, and convert the program audio channel number Location.   A program audio channel number conversion program for causing a computer to function as the program audio channel number conversion device according to any one of claims 1 to 6. In a broadcast program receiving apparatus for receiving a video signal and an audio signal of a broadcast program and reproducing the video signal and the audio signal,
A broadcast program receiver comprising the program audio channel number converter according to any one of claims 1 to 6.

Images