JP4381892B2 - Transmitter and receiver for sound quality correction transmission - Google Patents

Description

The present invention relates to a transmission device and a reception device in sound quality correction transmission, and more particularly to a transmission device and a reception device in sound quality correction transmission for providing high-accuracy sound by performing sound quality correction suitable for each individual.

  Conventionally, in a program, an effective production is realized by superimposing background sound (sound effect) such as BGM together with narration sound in a broadcast program. The audio in such a program is mixed and encoded with a certain predetermined mixing ratio, and an audio signal is provided to the viewer. For this reason, the viewer reproduces the audio signal decoded by the receiver or the like without correction, or performs reproduction using a preset sound quality correction characteristic. For this reason, it has been difficult to perform high-accuracy correction in consideration of the characteristics of each audio signal of the program and automatic sound quality correction suitable for personal preference.

  Therefore, as a conventional technique, there is a means for measuring the perceptual balance of sound with narration sound, background sound, and the like in program production, and controlling mixing using this as a control signal (see, for example, Patent Document 1). .

In addition, there is a method in which a viewer listens with a voice balance between “voice” and music or sound effects other than “voice” in a balance suitable for him / her (see, for example, Patent Document 2). In Patent Document 2, normal audio signals and audio signals other than human voice are assigned to different channels and multiplexed.
JP 2001-76460 A JP-A-10-327386

  However, in the case of the technique shown in Patent Document 1 described above, it is possible to select a preferred one from the sound produced in advance on the broadcast station side, but the sound produced on the broadcast station side is selected. It may not always be the best sound for each individual.

  In the case of the technique disclosed in Patent Document 2, two types of signals having different mixing balances between an electrical signal of a human voice and an electrical signal other than a human voice are received and the mixing balance is given to the viewer. Some adjustments can be made accordingly, but in this case as well, the sound may not be optimal for all people (viewers).

The present invention has been made in view of the above-described problems, and includes a transmitting device and a receiving device in sound quality correction transmission for performing sound quality correction suitable for each individual (each viewer) and providing high-accuracy sound. The purpose is to provide.

  In order to solve the above problems, the present invention employs means for solving the problems having the following characteristics.

The invention described in claim 1 transmits a program signal composed of a video and a voice having an audio signal including a narration sound and a background sound, and corresponds to viewer information including auditory measurement data for each viewer. A transmitting device that corrects the sound quality of the audio included in the program signal received on the receiving side, wherein either one of the narration sound and the background sound included in the audio signal is converted into an MDCT (Modified Discrete Cosine Transform). A maximum value of the total energy of the MDCT frequency data obtained by combining predetermined frequency bands selected from among a plurality of frequency bands constituting the frequency data of the converted audio signal is converted to frequency data. Generated as audio characteristic data of the band, and the generated audio characteristic data When the feature data generation means for multiplexing said voice, from the viewer information, the sound quality of the speech in order to make correction at the receiving side, the question including a questionnaire about the listening easiness of sound for each of the viewer Information and sound quality correction data generating means for generating sound quality correction data having audio information including an audio signal for measuring an individual's hearing associated with the question information, and obtained by the feature data generating means First transmission means for transmitting the program signal including the audio feature data to the reception side, and second transmission means for transmitting the sound quality correction data obtained by the sound quality correction data generation means to the reception side. The viewer information is obtained as a response to the sound quality correction data, and a desired mixing ratio of narration sound and background sound and a predetermined frequency band. The sound quality correction of the sound that is performed at the receiving side includes auditory measurement data that includes each hearing level, and corresponds to the hearing ratio for each predetermined frequency band from the viewer information. Correction data including a correction conversion table with a set volume level is generated, an audio signal including a narration sound and a background sound included in the sound is subjected to MDCT conversion, and the audio feature is obtained for each frequency band of the converted MDCT frequency data. Data is multiplied, and the MDCT frequency data obtained by the multiplication is multiplied or added / subtracted using the mixing ratio and the volume level for each viewer obtained corresponding to the frequency band of the correction conversion table. It is characterized by that.

According to the first aspect of the present invention, sound quality correction data for performing sound quality correction on the receiving side is generated in correspondence with the audio feature data, thereby realizing highly accurate sound quality correction on the receiving side efficiently. it can. Thereby, sound quality correction suitable for each individual can be performed, and highly accurate voice can be provided. Further, using the question information and the voice information associated with the question information, the setting by the viewer on the receiving side can be easily realized, and the sound quality correction data can be generated efficiently. In addition, it is possible to achieve a more effective effect by correcting each other's sound for the narration sound and the background sound that are mixed and output. Furthermore, audio feature data can be generated efficiently and with high accuracy.

The invention described in claim 2 receives the program signal transmitted from the transmission side, and converts the sound quality of the audio signal having the narration sound and the background sound in the received program signal into the viewer information for each viewer. Corresponding correction apparatus that receives and corrects the question information including a questionnaire regarding the ease of listening to the sound for each viewer included in the sound quality correction data transmitted from the transmission side, and associated with the question information For the audio information including the audio signal for measuring the individual hearing ability, the desired mixing ratio of the narration sound and the background sound and the hearing level for each predetermined frequency band, which are input as a response result from each viewer. from the viewing information obtaining means for obtaining audience information containing an auditory measurement data becomes, the auditory measurement data obtained by the viewing information obtaining means and By response, and sound quality correction data generation means for generating correction data consisting of the correction conversion table and volume level set corresponding to the hearing level and the mixing ratio for each of the predetermined frequency band, the sound quality correction data the correction data obtained by the generating means, to convert any one of the audio signals of the audio signal that includes a narration sound and background sound included in the audio and voice included in the program signal to the MDCT frequency data, it is converted Based on the audio feature data comprising the maximum value of the total energy of the MDCT frequency data obtained by combining predetermined frequency bands selected from among a plurality of frequency bands constituting the frequency data of the audio signal, the viewing have a correcting means for correcting the tone of the voice for each finisher, said correction means, the voice The audio signal including the narration sound and background sound included is subjected to MDCT conversion, the audio feature data is multiplied for each frequency band of the converted MDCT frequency data, and the MDCT frequency data obtained by the multiplication is stored in the correction conversion table. The sound quality of the sound is corrected by multiplying or adding / subtracting using the volume ratio for each viewer obtained corresponding to the mixing ratio and frequency band .

According to the second aspect of the present invention, the correction data is generated in advance from the sound quality correction data for performing the sound quality correction, so that the audio feature data can be efficiently and highly accurately corrected. Can do. Thereby, sound quality correction suitable for each individual can be performed, and highly accurate voice can be provided. Further, the viewer information can be easily obtained using the question information for the viewer and the voice information associated with the question information.

  ADVANTAGE OF THE INVENTION According to this invention, sound quality correction | amendment suitable for each individual can be performed and a highly accurate audio | voice can be provided.

<Outline of the present invention>
In order to perform sound quality correction suitable for each individual, the present invention is accompanied by a questionnaire regarding the ease of listening to sound by means of transmission / reception means using a communication network such as the Internet, a recording medium such as a CD-ROM, and the like. By acquiring an audio signal on the receiving device side and performing sound quality correction in accordance with the acquired information, personal sound quality correction data can be provided to each viewer.

  More specifically, the digital audio encoded stream signal accompanied by the audio signal and the characteristic data of the audio signal is received, for example, by digital broadcasting or Internet distribution, the received stream signal is decoded, and the decoded digital audio signal is decoded. In a device having a function of performing sound quality correction using the feature data included in the device, the sound quality correction suitable for each individual can be performed by combining the personal sound quality correction data of the individual acquired by this device and the feature data of the audio signal. Output the performed audio signal.

<Embodiment>
Embodiments to which the present invention is applied will be described below with reference to the drawings. FIG. 1 is a diagram showing a configuration example of a sound quality correction transmission system according to the present invention.

  A sound quality correction transmission system 10 illustrated in FIG. 1 is configured to include a transmission device 11 and a reception device 12. The transmission device 11 and the reception device 12 are connected in a state where data can be transmitted and received via a communication network 13 such as the Internet. Further, the transmission device 11 provides a broadcast program to the reception device 12 via the broadcast satellite 14.

  Here, the transmission device 11 shown in FIG. 1 is a transmission device composed of a provider or the like that provides a program or the like composed of video or audio that can be viewed by each viewer on the reception device 12 installed at home or the like. For example, a broadcast station or a content provider that provides content. In the present embodiment, the number of transmitting devices 11 and receiving devices 12 is not limited.

  Further, the transmission device 11 records the sound quality correction data such as the questionnaire for acquiring the viewer information and the test voice data on the recording medium 15 such as a CD-ROM, and provides it to the viewer terminal 12 on the reception device 12. can do.

  The transmission device 11 provides the produced broadcast program to the reception device 12 via the broadcast satellite 14. Further, the transmission device 11 transmits the sound quality correction data including the questionnaire to the viewer for causing the reception device 12 to correct the sound and the test sound data for measuring the viewing ability and the preference sound through the communication network 13. To the receiving device 12. The sound quality correction data described above can be stored in a recording medium 15 such as a CD-ROM and provided to the viewer by mail or the like.

  On the other hand, the receiving device 12 performs setting for each individual using question information such as a questionnaire from the transmitting device 11, data for sound quality correction, and the like, and generates sound quality correction data from information such as answers to the questionnaire and measurement results. Then, the generated sound quality correction data is accumulated to correct the sound included in the program.

  Next, functional configurations of the transmission device 11 and the reception device 12 will be described with reference to the drawings.

<Transmitter 11>
FIG. 2 is a diagram illustrating an example of a functional configuration of the transmission apparatus according to the present embodiment. The transmission device 11 shown in FIG. 2 includes an audio feature data generation unit 21, a program production unit 22, a sound quality correction data generation unit 23, a first transmission unit 24, and a second transmission unit 25. ing.

  The audio feature data generation means 21 receives narration audio data and background sound audio data such as BGM, and generates audio feature data based on the genre information of the broadcast program or the audio signal of each program. Further, the audio feature data generation unit 21 outputs the generated audio feature data to the program production unit 22. The detailed description of the audio feature data generation will be described later.

  The program production means 22 produces the program by inputting the video data and the audio feature data obtained from the audio feature data generation means 21. Further, the program production means 22 outputs the produced program signal to the first transmission means 24.

  Further, the sound quality correction data generation means 23 generates the sound quality correction data by the receiving device 12 and performs sound quality correction, so that the question information for the viewer such as a questionnaire and the voice information (test voice) associated with the question information. Sound quality correction data is generated from the data. The sound quality correction data generation unit 23 outputs the generated sound quality correction data to the second transmission unit 25. Note that the sound quality correction data generating means 23 can store the sound quality correction data generated on the recording medium 15 such as a CD-ROM, thereby sending the sound quality correction data to the receiving device 12 by mail or the like. be able to.

  Here, the question information is, for example, a questionnaire regarding ease of listening to the viewer and data regarding various questions, and the test voice data is, for example, an audio signal for measuring an individual's hearing associated with the questionnaire. It is. Also included are sample data of narration sounds and background sounds.

  The first transmission means 24 outputs the program signal obtained from the program production means 22 to the receiving device 12 by the broadcast satellite 14. The second transmission means 25 outputs the sound quality correction data obtained from the sound correction data generation unit 23 to the receiving device 12 via the communication network 13 or the like.

  Here, the signal transmitted from the first transmitting unit 24 is transmitted for each program generated by the program producing unit 22, whereas the signal transmitted from the second transmitting unit 25 is used to generate audio correction data. When the reception apparatus 12 performs correction using the sound quality correction data generated by the unit 23, for example, transmission is performed only once.

<Audio Feature Data Generation Unit 21: Audio Feature Data Generation Example>
Next, an example of generating audio feature data in the audio feature data generating unit 21 will be described with reference to the drawings. FIG. 3 is a diagram for explaining an example of generation of audio feature data. Specifically, an example of a generation processing block that generates audio feature data from a narration sound is shown.

  Here, for example, in an information providing program such as news, an example of generation of audio feature data for making the narration sound clear and easy to hear with respect to the narration sound and the background sound will be described. The audio sound used in this embodiment is not limited to this, and is not limited to two audio signals (narration sound and background sound).

  Further, here, feature data corresponding to 49 auditory frequency bands (Scalefactor Bands) encoded by MPEG-2 AAC (MPEG: Moving Picture Experts Group / AAC: Advanced Audio Coding) extracted from narration sound is generated. Shall.

  The audio feature data generating means 21 shown in FIG. 3 includes a mixing means 31, an MDCT (Modified Discrete Case Transform) means 32, an MPEG-2 AAC means 33, a scale factor band combining means 34, and a scale factor band maximum value detecting means. 35, an encoding unit 36, and a multiplexing unit 37.

  Here, normally, the narration sound and the background sound are input separately in the transmission device 11 and mixed with an appropriate level ratio by the mixing means 31 to generate broadcast sound. In FIG. 3, audio feature data is generated using only the narration sound.

  First, the narration voice is input to the MDCT means 32, and the MDCT window information obtained by encoding the voice by the MPEG-2 AAC encoding means 33 is input to convert the time signal into MDCT frequency data.

  FIG. 4 is a diagram showing a frequency band and a range of MDCT frequency data. In FIG. 4, the sample rate fs is 44.1 kHz or 48 KHz, and the MDCT frequency data range (swb_offset_long_window) in each of the scale factor bands (swb) of 0 to 48 in 49 auditory frequency bands (num_swb_long_window) is shown. Yes.

  Here, for example, as the audio feature data of the narration sound to be multiplexed, the data up to the 35th (swb = 35) of the scale factor band (swb) shown in FIG. 4 among the converted MDCT frequency data is used. And

  That is, the data frequency-converted by the MDCT unit 32 is combined by the scale factor band combining unit 34, and the total energy of the MDCT frequency data in the scale factor band shown in FIG. 4 is calculated. Further, the scale factor band combining unit 34 outputs, for example, a signal combined up to the 34th to the scale factor band maximum value detecting unit 35. The scale factor band maximum value detecting means 36 outputs the maximum value of the total energy of the MDCT frequency data in the input signal to the encoding means 36 as scale factor band feature data.

  The encoding unit 36 encodes the input feature data and outputs it to the multiplexing unit 37. Multiplexing means 37 multiplexes the digital audio stream and the audio feature data and outputs the result. Note that, when it is determined that the narration sound is sufficiently loud, the feature data is multiplexed. However, when it is determined that the narration sound is close to silence, it is not necessary to multiplex the feature data. Next, an example of a multiplexing method will be specifically described.

<Example of multiplexing feature data into audio stream>
Here, an example of a method of multiplexing feature data into an audio stream by the audio feature data generation means 21 will be described with reference to the drawings. Here, an example of a method for multiplexing audio feature data associated with a digital audio encoded stream will be described by taking MPEG-2 AAC encoding as an example.

  FIG. 5 is a diagram illustrating a syntax example of the raw_data_block syntax. As shown in FIG. 5, the audio data of the MPEG-2 AAC digital audio encoded stream is multiplexed in the form of raw_data_block syntax as shown in FIG. In this syntax, a fill_element area and a data_stream_element area are defined. Here, each of the above-described regions will be described.

  FIG. 6 is a diagram illustrating a syntax example of the fill_element area, and FIG. 7 is a diagram illustrating a syntax example of the data_stream_element area. Further, FIG. 8 is a diagram illustrating a syntax example of the extension_payload area included in the fill_element area illustrated in FIG.

  As shown in FIG. 6, the audio feature data is multiplexed as the acoustic feature_extension_data () shown in the extension_payload syntax of FIG. 8 included in the fill_element region, or as the acoustic feature_extension_data_byte_element shown in the data_stream_element region shown in FIG. Can be multiplexed.

  In addition, in order to multiplex as extension_payload syntax acoustic feature_extension_data (), it seems necessary to be approved by standards such as MPEG, but when multiplexing as acoustic feature_extension_data_byte shown in the data_stream_element area, MPEG Therefore, multiplexing can be easily realized. The example of multiplexing feature data in the present invention is not limited to this.

<Receiver 12>
Next, a specific functional configuration of the receiving device 12 will be described with reference to the drawings. FIG. 9 is a diagram illustrating an example of a functional configuration of the receiving device according to the present embodiment. The receiving apparatus 12 shown in FIG. 9 includes a receiving means 41, a sound quality correction data generating means 42, a sound quality correction data accumulating means 43, a video / audio separating means 44, a sound quality correcting means 45, a viewer information means 46, The display unit 47, the audio output unit 48, and the input unit 49 are configured.

  The receiving means 41 receives the program signal from the broadcast satellite 14 and outputs it to the video / audio separating means 44. In addition, the receiving unit 41 acquires various types of information on the sound quality correction data transmitted from the transmission device 11 via the communication network 13. Here, the various pieces of information acquired are question information and test audio information for correcting the audio for the viewer described above. The receiving unit 41 outputs the acquired sound quality correction data to the viewer information acquiring unit 46.

  The sound quality correction data generating means 42 performs sound quality correction corresponding to the audio feature data included in the digital audio stream by using the answer result and the individual auditory measurement data in the questionnaire acquired by the viewer information acquiring means 46. An audio correction conversion table that can be generated is generated. Further, the sound quality correction data generating means 42 outputs the generated conversion table to the sound quality correction data accumulating means 43. The sound quality correction data storage means 43 stores the input sound correction conversion table for each viewer (individual).

  The video / audio separation unit 44 separates the input program signal into a video signal and an audio signal, outputs the separated video signal to the display unit 47, and outputs the separated audio signal to the sound quality correction unit 45.

  The sound quality correction means 45 uses the feature data included in the decoded digital audio signal and the sound quality correction conversion table of each viewer stored in the sound quality correction data storage means 43, so that the sound quality correction optimum for the individual according to the program is performed. Is generated and output to the audio output means 48. This output is, for example, an audio signal in which the narration sound is emphasized. Details of the sound quality correction will be described later.

  The viewer information acquisition means 46 inputs various kinds of sound quality correction information obtained from the reception means 41. Note that, as described above, these pieces of information are question information and test audio information for correcting the audio for the viewer, and can also be acquired from the recording medium 15 or the like.

  The viewer information acquisition device 46 outputs the input question information to the display means 47. Also, the test voice information is output to the voice output means 47. In addition, the viewer information acquisition device 46 receives the answer of the question output by the viewer to the display means 47 by the input means 49 such as a keyboard, a mouse, or a remote controller, the hearing measurement data of each individual, setting information such as volume, etc. get. Further, the viewer information acquisition device 46 outputs the acquired data to the sound quality correction data generation means 42.

  The display means 47 is composed of a monitor or the like and outputs the video obtained from the sound separation means 44. In addition, contents such as a questionnaire regarding voice correction are displayed to the viewer input from the viewer information acquisition means 46.

  The audio output means 48 is composed of a speaker or the like, and the sound quality correction means 45 outputs an audio signal suitable for the viewer's preference. Also, test audio data input from the viewer information acquisition means 46 is output.

  The input means 49 includes a keyboard, a mouse, a remote controller, etc., and the viewer can store the sound quality correction data of a plurality of viewers with one receiving device by the input means 49. Enter personal type information for identifying Further, the question information and test voice data are auditioned and the answer is input. For example, an answer is given to a question such as whether the audio signal of the test voice data is heard or easy to hear.

  In addition, when viewing the video, the program, etc., the input unit 49 causes the viewer information acquisition unit 46 to perform processing related to viewer information acquisition when it is desired to newly generate correction data based on the above-described question information and test audio data. Can be instructed to do.

<Example of viewer information acquisition>
Here, a specific audio setting example for acquiring the viewer information will be described.

  First, the sine wave sound (for example, 500 Hz) is reproduced at a predetermined output level by the audio output means 48. The viewer answers “Yes” using the input means 49 if the reproduced sound is heard, and “No” if the reproduced sound is not heard. This answer result is acquired by the viewer information acquisition means 46.

  Next, if the reproduction sound is heard from the answer result, the output level is lowered, and if the reproduction sound is not heard, the output level is raised and reproduction is performed again. The above-described processing is repeated, and a threshold value that the sound cannot be heard below a certain level but can be heard above a predetermined level is acquired as auditory data for a certain sine wave sound of the respondent.

  Next, by using a sine wave sound (for example, 1000 Hz) different from the above sine wave sound, it is acquired as auditory data for a certain sine wave sound of the respondent individual as described above. In this way, the above-described processing is repeated for all of the plurality of sine wave sounds to be measured.

  Next, a program sound (Ref) simulating a broadcast program and a program sound (Tst) in which a narration sound and a sound effect are set at a certain ratio are reproduced from the audio output means 48. If the voice output means 48 prefers Tst over Ref, “Yes” is answered, and if Ref is preferred over Tst, “NO” is answered, and the answer information is obtained by the viewer information obtaining means 46.

  Next, if Tst is preferred over Ref as a result of the answer, the background sound ratio is decreased, and if Ref is preferred over Tst, the background sound ratio is increased and reproduction is performed again. The above process is repeated, and a threshold value that favors Tst over Ref below a certain ratio and prefers Ref over Tst above a certain ratio is acquired as preferred auditory data for a certain program sound. Further, the above-described processing is repeated for all the program sounds to be measured.

<Sound quality correction data generating means 42>
Next, a specific example of table generation in the sound quality correction data generation means 42 will be described. In the above-described viewer information acquisition example, when a predetermined number of data can be acquired from the acquisition data acquired by the viewer information acquisition means 46, a table is generated from the acquired data.

  Specifically, the collected hearing level for each individual frequency (the lowest level that can be heard for each individual) and the data of the desired mixing ratio between the narration for each frequency and the background sound are used as the sound quality correction conversion table. . Alternatively, the sound quality correction conversion table is generated by creating data of the hearing level for each frequency and the desired mixing ratio between narration and background sound for each sound volume. Note that the table generation example is not limited in the present invention, and other methods may be used.

  Here, a more specific conversion table will be described. Here, an example of creating a gain conversion table from auditory data for the acquired sine wave sound will be described.

  FIG. 10 is a diagram illustrating an example of the gain conversion table. In FIG. 10, the gain conversion table has a value of 0 to 255 as a scale factor value (sfb value) in MPEG-2 AAC for each scale factor band number (sfb number) as in the audio feature data generation example described above. Therefore, a gain conversion table in which predetermined gain values (A to ZZ48) are set for a two-dimensional array of sfb numbers and sfb values is generated.

  FIG. 11 is a diagram illustrating an example of data transition before and after correction. Here, the predetermined gain value in the gain conversion table shown in FIG. 10 can be a difference value before and after gain correction shown in FIG. Thus, according to FIG. 10, a different gain value can be described for each scale factor band.

  Here, as shown in FIG. 11, when it is difficult to hear a sound smaller than a certain input volume in the auditory data collected from the viewer, the predetermined range (50 to 100) is corrected to a constant volume. For a volume exceeding a certain predetermined amount (200 or more in FIG. 11), correction is performed so that the slope of the output with respect to the input is halved.

  In other words, by performing the correction as described above, for example, the program sound of a program (such as an information providing program such as news) broadcasted to all receiving devices is broadcast by another receiving device using correction data. Sounds with narration greater than sound can be easily output.

  In addition, when generating a desired mixing ratio for program sound from the auditory measurement data, first, the viewer sets a desired sound quality threshold value based on the above-described auditory measurement data for sine wave sound. It is possible to generate a table or the like for making the volume difference (dynamic range) between the smallest sound and the largest sound smaller than the broadcast sound.

  Also, it is possible to generate a mixing ratio table for changing the mixing ratio of narration sound and background sound such as sound effect from broadcast sound. In this mixing ratio table, audio feature data is included in a digital audio encoded stream, and audio feature data related to a narration sound is transmitted in an information providing program such as news in order to make the narration sound clear and easy to hear. Used in cases.

  As shown in FIG. 10, the mixing ratio table has a scale factor value (sfb value) of 0 to 255 in MPEG-2 AAC for each scale factor band number (sfb number) as in the audio feature data generation example described above. Since there is a value, a mixing ratio in which a predetermined gain value (A to ZZ48) is set can be set for a two-dimensional array of sfb numbers and sfb values.

  Here, for example, when the collected auditory data is a single value of an information providing program such as news, a constant mixing ratio can be set regardless of the sfb value. In addition, when used for information providing programs such as news, the mixing ratio up to sfb number 35 is corrected in accordance with the collected auditory data, and the mixing level ratio of 35 or higher is set to be the same mixing ratio as normal audio. You can also.

<Sound quality correction means 45>
Next, the sound quality correction means 45 will be specifically described. The sound quality correction means 45 uses the personal sound quality correction conversion table to combine the acquired personal sound quality correction data with the genre information of the broadcast program or the feature data included in the audio signal for each program and transmitted. The sound quality is corrected so that an audio signal suitable for the viewer is obtained for each program.

  FIG. 12 is a diagram illustrating an example of a correction processing block in the sound quality correction unit. FIG. 12 is a block configuration example for making the narration sound clear and easy to hear. 12 includes a separating unit 61, a digital audio stream decoding unit 62, a narration decoding unit 63, an MDCT unit 64, a narration mix ratio conversion unit 65, a dynamic range conversion unit 66, and an inverse MDCT unit 67. It is configured as follows.

  The block diagram shown in FIG. 12 shows a case where the digital audio signal is once decoded and the sound quality correction is performed on the decoded signal. However, in order to simplify the configuration, this function may be incorporated when decoding the digital audio stream. Is possible.

  Separation means 61 separates the input MPEG-2 AAC digital audio stream and narration feature data. The separating unit 61 outputs the separated digital audio stream to the digital audio stream decoding unit 62 and outputs the narration feature data to the narration feature data decoding unit 63.

  The digital audio stream decoding unit 62 decodes the digital audio stream obtained by the separating unit 61 into a digital audio signal. Further, the digital audio stream decoding unit 62 outputs the decoded digital audio signal to the MDCT unit 64. The narration feature data decoding unit 63 decodes the narration feature data. The narration feature data decoding unit 63 outputs the decoded narration feature data to the MDCT unit 64.

  The MDCT means 64 converts the time signal into MDCT frequency data again using the same MDCT window information as the frame of the MPEG-2 AAC digital audio stream.

  Specifically, the MDCT frequency data obtained by the MDCT means 64 is multiplied by the narration feature data for each scale factor band to reproduce the MDCT frequency data of the narration sound, and the MDCT frequency data of the reproduced narration sound is converted to the narration mix ratio. It outputs to the conversion means 65.

  The narration mix ratio conversion means 65 uses the MDCT frequency data of the original broadcast sound (mixed narration sound and background sound) as data of the sound quality correction table (mixing ratio table) obtained from the sound quality correction data storage means 43. Accordingly, the sound quality is corrected by multiplying or adding / subtracting with a mixing ratio appropriate for the viewer.

  The narration mix ratio conversion means 65 outputs the sound whose sound quality has been corrected to the dynamic range conversion means 66. In accordance with the gain conversion table, the dynamic range conversion unit 66 multiplies or adds / subtracts the input signal with the data of the sound quality correction table (gain conversion table) obtained from the sound quality correction data storage unit 43, so that the volume level can be viewed. Create a sound quality corrected signal that falls within the appropriate range for the user. The dynamic range conversion unit 66 outputs the signal to the inverse MDCT unit 67.

  The inverse MDCT means 67 performs inverse MDCT on the signal obtained by the dynamic range conversion means 66 and outputs a sound quality corrected signal. Thereby, sound quality correction suitable for each individual can be performed, and highly accurate voice can be provided.

  As described above, sound quality correction suitable for each individual can be performed to provide highly accurate sound.

  According to the present invention, in the form of broadcasting or the like in which a large number of unspecified persons are simultaneously auditioned, the transmitting side transmits the same program having characteristic data to a plurality of receiving sides, and the receiving side has the optimum sound quality for each individual (for each viewer) The user who wants to listen to the sound generates sound quality correction data by answering based on the sound quality correction data consisting of the question information such as questionnaire about the ease of hearing of the sound obtained in advance and the test voice data attached to the question information In addition, by combining the generated sound quality correction data and the audio feature data included in the digital broadcast wave or digital audio encoded stream signal of Internet sound, sound quality correction suitable for each individual is performed and high-accuracy audio is provided. Can do.

  For example, broadcast programs include various genres such as news programs, information programs such as news commentary, sports broadcasts, and music programs. Of these, it is desirable for information providing programs such as news that the narration voice is clear and easy to hear and the contents can be accurately grasped. On the other hand, in music programs, it is important to maintain the atmosphere of the program sound. In addition, hearing ability and preference for each individual are different, and it is necessary to correct sound quality appropriately for each individual. The present invention can provide an audio signal that has been subjected to optimum sound quality correction for an individual in accordance with a program. By applying the present invention, it is possible to make it easy to hear lines, announcements, and narrations of programs and the like that are difficult for the elderly and the like to hear.

  Although the preferred embodiment of the present invention has been described in detail above, the present invention is not limited to the specific embodiment, and various modifications, within the scope of the gist of the present invention described in the claims, It can be changed.

It is a figure which shows the example of 1 structure of the sound quality correction | amendment transmission system in this invention. It is a figure which shows an example of a function structure of the transmitter in this embodiment. It is a figure for demonstrating the example of a production | generation of audio feature data. It is a figure which shows the range of Scalefactor bands and MDCT frequency data. It is a figure which shows the syntax example of raw_data_block syntax. It is a figure which shows the syntax example of a fill_element area | region. It is a figure which shows the example of a syntax of a data_stream_element area | region. It is a figure which shows the syntax example of the extension_payload area | region contained in a fill_element area | region. It is a figure which shows an example of a function structure of the receiver in this embodiment. It is a figure which shows an example of a gain conversion table. It is a figure which shows an example of the data transition before correction | amendment and after correction | amendment. It is a figure which shows an example of the correction process block in a sound quality correction | amendment means.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Sound quality correction | amendment transmission system 11 Transmission apparatus 12 Reception apparatus 13 Communication network 14 Broadcasting satellite 15 Recording medium 21 Audio characteristic data generation means 22 Program production means 23 Sound quality correction data generation means 24 1st transmission means 25 2nd transmission means 31 Mixing means 32 MDCT means 33 MPEG-2 AAC encoding means 34 Scale factor band combining means 35 Scale factor band maximum value detecting means 36 Encoding means 37 Multiplexing means 41 Transmitting means 42 Sound quality correction data generating means 43 Sound quality correction accumulating means 44 Video / audio Separation means 45 Sound quality correction means 46 Viewer information acquisition means 47 Display means 48 Audio output means 49 Input means 61 Separation means 62 Digital audio stream decoding means 63 Narration feature data decoding means 64 MDCT hand 65 narration mix ratio converting means 66 dynamic range conversion means 67 inverse MDCT unit

Claims (2)

Video, performs transmission of the program signal comprising a voice having an audio signal including a narration sound and background sound, in association with the audience information containing the auditory measurement data for each viewer, the program signal received by the receiving side A transmission device for correcting the sound quality of the sound included in
Any one of the narration sound and background sound included in the audio signal is converted into MDCT (Modified Discrete Case Transform) frequency data, and selected from a plurality of frequency bands constituting the frequency data of the converted audio signal Characteristic data for generating a maximum value of the total energy of the MDCT frequency data obtained by combining predetermined frequency bands to be generated as audio characteristic data of the frequency band, and multiplexing the generated audio characteristic data and the voice Generating means;
From the viewer information , in order to correct the sound quality of the sound on the receiving side , question information including a questionnaire regarding the ease of listening to the sound for each viewer, and the individual's hearing associated with the question information Sound quality correction data generating means for generating sound quality correction data having audio information including an audio signal for measuring
First transmission means for transmitting a program signal including audio feature data obtained by the feature data generation means to the reception side;
Have a second transmitting means for transmitting a sound quality correction data obtained by the sound quality correction data generation means on the receiving side,
The viewer information includes auditory measurement data, which is obtained as a response result to the sound quality correction data, and includes a desired mixing ratio of the narration sound and the background sound and a hearing level for each predetermined frequency band,
The sound quality correction performed on the receiving side is a correction composed of a correction conversion table of the mixing ratio and the volume level set corresponding to the hearing level for each predetermined frequency band from the viewer information. MDCT frequency obtained by generating data, performing MDCT conversion on an audio signal including narration sound and background sound included in the sound , multiplying the audio feature data for each frequency band of the converted MDCT frequency data, and multiplying A transmission apparatus characterized by multiplying or adding / subtracting data using the mixing ratio of the correction conversion table and the volume level for each viewer obtained corresponding to the frequency band . A receiving device that receives a program signal transmitted from a transmission side and corrects the sound quality of an audio signal having a narration sound and a background sound in the received program signal in accordance with viewer information for each viewer. ,
Question information including a questionnaire regarding the ease of listening to sound for each viewer included in the sound quality correction data transmitted from the transmission side, and an audio signal for measuring an individual's hearing associated with the question information Viewer information including auditory measurement data, which is input as an answer result from each of the viewers, and includes a desired mixing ratio of the narration sound and the background sound and a hearing level for each predetermined frequency band Viewer information acquisition means for acquiring
Corresponding to the auditory measurement data obtained by the viewer information acquisition means, correction data comprising a correction conversion table of the mixing ratio and the volume level set corresponding to the hearing level for each of the predetermined frequency bands Sound quality correction data generating means for generating
Wherein the correction data obtained by the tone correction data generating means converts the one of the audio signals of the audio signal that includes a narration sound and background sound included in the audio and voice included in the program signal to the MDCT frequency data Based on audio feature data comprising a maximum value of total energy of the MDCT frequency data obtained by combining predetermined frequency bands selected from among a plurality of frequency bands constituting frequency data of the converted audio signal Te, have a correction means for correcting the sound quality of the voice for each of the viewer,
The correction means includes
The audio signal including the narration sound and background sound included in the voice is subjected to MDCT conversion, the audio feature data is multiplied for each frequency band of the converted MDCT frequency data, and the MDCT frequency data obtained by the multiplication is corrected. A receiving apparatus , wherein the sound quality of the sound is corrected by multiplying or adding / subtracting using the mixing ratio of the conversion table and the volume level for each viewer obtained corresponding to a frequency band .

Images