JP5106889B2 - Audio output device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a voice output apparatus which adjusts gain of output voice according to surrounding noise, without depending on a hardware structure of a voice output stage. <P>SOLUTION: In an output control section 21 of a voice output application section 2, regarding each divided voice data in which voice data D are divided for each time length Ts, a voice data PD to which gain adjustment according to a current volume level of surrounding noise is sequentially performed on a divided voice data for each time interval of the time length Ts, by using a gain adjustment section, is generated, while HDR in which an address and a size of the PD are indicated, is informed, and thereby, processing for requesting outputting of voice which generated voice data express, to a sound driver (output) 11, is performed. The sound driver (output) 11 sequentially outputs voice expressed by each voice data indicated by informed HDR express, from a speaker 6 via a sound output device 5. <P>COPYRIGHT: (C)2008,JPO&amp;INPIT

Description

  The present invention relates to an audio output device that dynamically changes an output audio gain according to a change in ambient noise so that a user's audio perception is well maintained.

As an audio output device that dynamically changes the frequency gain characteristics of the output audio in accordance with changes in the ambient noise so that the user's audio perception is well maintained, the audio output in accordance with the ambient noise A device that adjusts the gain characteristic of this is known (for example, Patent Document 1).
JP 2004-23481 A

The adjustment of the gain characteristic of the output sound in accordance with the ambient noise in the sound output apparatus as described above is performed by performing gain adjustment in real time on the output sound while outputting the sound. It has been broken.
Here, in order to perform gain adjustment in real time while outputting sound in this way, it is necessary to perform the gain adjustment by hardware or software at the sound output stage. However, in a device that does not include a special element or circuit that performs gain adjustment according to ambient noise, the gain adjustment cannot be performed by hardware. On the other hand, in order to realize the gain adjustment by software, for each type of hardware in the audio output stage, prepare software that performs gain adjustment according to the ambient noise by processing according to the hardware configuration. Need to do.

  Therefore, an object of the present invention is to provide a configuration of an audio output device that can realize gain adjustment of an output audio according to ambient noise regardless of the hardware configuration of the audio output stage.

  In order to achieve the above object, the present invention provides a voice output device that outputs voice by reading out voice data stored in the memory and designated as output voice data in an issued voice output request. A voice output unit that outputs a voice represented by the read voice data, a noise level detection unit that detects a volume level of ambient noise, and target voice data that is voice data that represents the voice to be output. Dividing into divided voice data that is voice data representing voice for each section, each divided voice data is sequentially set as target divided voice data every predetermined time, and the voice represented by the target divided voice data is Generating gain-adjusted audio data representing audio amplified with a gain corresponding to the volume level of ambient noise detected by the noise level detection means, Audio output processing means for storing the adjusted audio data in the memory and issuing the audio output request specifying the stored gain-adjusted audio data as the output audio data to the audio output unit It is.

  According to such an audio output device, since the audio data itself supplied to the audio output unit that outputs the audio is subjected to gain adjustment according to the ambient noise by the audio output processing unit, the audio output unit Regardless of the configuration, it is possible to achieve gain adjustment of output sound in accordance with ambient noise.

  Here, when the audio data itself supplied to the audio output unit that outputs audio is subjected to gain adjustment according to ambient noise, the time when the gain adjustment is performed on the audio data, and the audio The time difference that occurs between the time when the voice represented by the data is output becomes a problem. This is because the ambient noise considered in the gain adjustment performed on the output sound becomes the time difference in the past.

  However, according to the present audio output device, in the audio output processing means, the gain adjustment of the audio data and the supply of the audio data to the audio output unit are divided by dividing the target audio data that is the audio data representing the audio to be output. Since audio data can be used as a unit, such a time difference can be shortened, and an audio with gain adjusted according to the current ambient noise condition can be output. In addition, since it is considered that the ambient noise in the adjacent time is approximated, the sound that has been gain-adjusted in this way in the divided sound data unit can be obtained by sufficiently shortening the time length of the sound represented by the divided sound data. Practically, it can be used as a sound that has been gain-adjusted according to the current ambient noise situation.

  More specifically, such an audio output device outputs a microphone, a sound input device that outputs ambient audio data representing ambient audio picked up using the microphone, a memory, and audio. The sound output device accepts the issued voice output request, reads the voice data stored in the memory specified as the output voice data in the accepted voice output request, and reads the voice represented by the read voice data A sound driver to be output to the sound output device; noise level detection means for detecting a volume level of ambient noise represented by the ambient audio data output by the sound input device; and target audio data that is audio data representing the audio to be output. , Divided into divided audio data that is audio data representing audio for each time interval of a predetermined time length. Each divided voice data is sequentially set as target divided voice data at each predetermined time, and the voice represented by the target divided voice data is determined according to the volume level of ambient noise detected by the noise level detecting means at the time. Generating the sound data after gain adjustment representing the sound amplified by the gain, storing the generated sound data after gain adjustment in the memory, and specifying the stored sound data after gain adjustment as the output sound data You may make it comprise as an audio | voice output apparatus provided with the audio | voice output process means which issues a request | requirement to the said sound driver.

When configured in this way, it is possible to realize gain adjustment of the output sound almost in accordance with the current ambient noise, regardless of the sound output device or the sound driver.
Here, each of the above audio output devices always holds the ambient audio data representing the ambient audio for the predetermined past length of time in the past until at least the audio output request is issued. If the ambient noise holding means and the voice output request are issued, at that time, the volume level of ambient voice data for the predetermined time length in the latest past held in the ambient noise holding means is Calculated as the volume level of the ambient noise to be detected, and thereafter, for each passage of the time period of the predetermined time length, the volume level of the ambient sound in the time period of the predetermined time length that has passed most recently A volume level calculating means for calculating the volume level of noise, and if the audio output request is issued in the audio output processing means, the volume level is calculated. Each time the stage calculates the volume level of the ambient noise, each divided voice data is sequentially set as target divided voice data, and the voice represented by the target divided voice data is set according to the calculated volume level of the ambient noise. It is also possible to generate gain-adjusted audio data representing the audio amplified by the gain.

  In this way, as soon as an audio output request is issued, the noise level detection means immediately uses the ambient audio data for the predetermined time length in the most recent past, which is always held in the noise level detection means. The level can be detected. Accordingly, immediately after the generation of the audio output request, the audio output processing means can generate and store the gain-adjusted audio data and issue the audio output request, and can immediately output the audio.

  In the above sound output device, the noise level detection means detects the volume level of ambient noise for each frequency band, and the sound output processing means detects the sound represented by the target divided sound data for each frequency band. In addition, it may be configured to generate gain-adjusted sound data representing sound amplified by a gain corresponding to the volume level of each frequency band of ambient noise detected by the noise level detection means at the time.

  Moreover, the said audio | voice output apparatus mentioned above may be mounted in a motor vehicle.

  As described above, according to the present invention, the gain adjustment of the output sound according to the ambient noise can be realized regardless of the hardware configuration of the sound output stage.

Hereinafter, embodiments of the audio output device according to the present invention will be described taking application to an in-vehicle audio output device as an example.
First, the first embodiment will be described.
FIG. 1 shows the configuration of the audio output device.
As shown in the figure, the audio output device includes an operating system 1, an audio output application 2, a sound input device 3, a microphone 4, a sound output device 5, a speaker 6, an audio data memory 7, and an output buffer memory 8. Yes.
However, the audio output apparatus as described above has a general electronic computer configuration including a CPU, a memory, an external storage device, and the like as a hardware configuration. Further, as the sound input / output hardware of the electronic computer, the sound input device 3, the microphone 4, the sound output device 5, and the speaker 6 are provided. The operation system, the audio output application 2, the audio data memory 7, and the output buffer memory 8 described above are realized as processes and storage resources on the electronic computer by executing a program prepared in advance by the CPU. It is what is done.

  In the configuration of such an audio output device, the operating system 1 includes a sound driver (input) 12. Then, the sound driver (input) 12 generates sound data representing ambient noise picked up by the microphone 4 and taken in via the sound input device 3.

  In addition, the operating system 1 includes a sound driver (output) 11 that controls the sound output device 5 and outputs sound to the speaker 6. Here, the sound driver (output) 11 includes an HDR queue that stores an HDR representing an address and a size of audio data of audio to be output. Then, the sound driver (output) 11 sequentially extracts the HDR stored in the HDR queue, reads out the audio data for the size indicated by the HDR from the address indicated by the extracted HDR, and outputs the audio represented by the audio data to the sound output device 5. The process which outputs to the speaker 6 via this is performed. In addition, when the sound driver (output) 11 finishes outputting the audio represented by the audio data indicated by the HDR as described above, the sound driver (output) 11 notifies the reproduction end notification to the issuer of the HDR stored in the HDR queue.

The audio output application 2 includes an output control unit 21 and a gain adjustment unit 22.
Hereinafter, the operation of the audio output application 2 will be described.
When the audio data D representing the audio to be output from the speaker 6 stored in the audio data memory 7 is generated on the audio data memory 7, the output control unit 21 of the audio output application 2 targets the audio data D. Perform audio output processing. Here, the voice data D may be voice data stored in the voice data memory 7 in advance, or may be newly generated by voice synthesis processing or the like. The voice data D represents, for example, voice for guidance to the user. In this case, the time length of the voice represented by the voice data D is several seconds to several tens of seconds.

FIG. 2 shows a procedure of audio output processing performed by the output control unit 21 of the audio output application 2.
As shown in the figure, in this processing, first, various preprocessing such as declaration of start of use of the device (sound output device 5) and declaration of format of audio data to be output is performed on the sound driver (output) 11. (Step 202).
Next, the following processing is sequentially performed on the divided audio data D (j) corresponding to the number n of audio data divisions obtained by dividing the audio data D, starting from the smallest j (steps 204, 220, and 228). Here, the audio data D is divided from j × ts (j + 1) of the audio represented by the audio data D, with the audio represented by each divided audio data D (j) having ts as a predetermined time length. ) Xts so as to represent the voice in the period. However, the audio time length represented by the audio data D is L, the audio data division number n is determined as the smallest integer satisfying L ≦ n × ts in step 204, and j is an integer satisfying 0 ≦ j <n. .

  Next, a remainder with j of 2 as a modulus is obtained as i (step 206), and it is checked whether j is less than 2 (step 208). If it is less than 2, the gain adjustment unit 22 sends the divided audio data D. (J) is subjected to gain adjustment according to the ambient noise at the present time, and the audio data obtained by adjusting the gain of the divided audio data D (j) is stored as PD (i) in the output buffer memory 8 (step 214). . However, when j is 1, it is preferable to start the execution of the current step 214 after the time length ts has elapsed since the execution of the previous step 214 was started.

  Then, the address and size attributes of PD (i) are set to HDR (i) (step 216), HDR (i) is issued as HDR to the sound driver (output) 11 and stored in the HDR queue (step 218). ). Here, the gain adjustment unit 22 of the audio output application 2 acquires ambient noise audio data from the sound driver (input) 12 in response to a request from the output control unit 21, and the ambient noise volume level represented by the acquired audio data Is applied to the divided audio data D (j). That is, the divided audio data that stores the audio data representing the audio represented by the divided audio data D (j) with the amplification factor corresponding to the volume level of the ambient noise is stored as PD (i) in the output buffer memory 8. Data D (j) is generated as audio data with gain adjustment. The gain adjustment of the gain adjusting unit 22 is performed by obtaining the volume level of the ambient noise represented by the sound data of the ambient noise for each frequency band, and adjusting the gain adjustment according to the volume level of each obtained frequency band of the ambient noise. You may make it carry out by giving to division | segmentation audio | voice data D (j) for every zone | band.

  Next, if j is not less than 2 (step 208), the process waits for a reproduction end notification from the sound driver (output) 11 (step 210), clears HDR (i) (step 212), and then divides. The audio data D (j) is subjected to gain adjustment according to the ambient noise at the present time, and the audio data obtained by adjusting the gain of the divided audio data D (j) is stored as PD (i) in the output buffer memory 8 ( Step 214).

Then, the address and size attributes of PD (i) are set to HDR (i) (step 216), HDR (i) is issued as HDR to the sound driver (output) 11 and stored in the HDR queue (step 218). ).
When the above processing is completed for all the divided audio data D (j) obtained by dividing the audio data D, the reproduction of the sound driver (output) 11 is awaited (step 222), and HDR (0) is set. HDR (1) is cleared (step 224), and post-processing such as declaration of the end of use of the device (sound output device 5) is performed on the sound driver (output) 11 (step 226), and audio output processing is performed. finish.

FIG. 3 shows an example of such audio output processing.
Now, as shown in FIG. 3a, when the audio data D represents the audio for a time length of 6 × ts, the audio data D is divided into divided audio data from D (0) to D (5). Is done.
In this case, as shown in FIG. 3b, in the first time interval t0 from the time interval t0 to t6 for each time length ts, the first divided audio data D (0) of the audio data D is The audio data GD (0) gain-adjusted according to the volume level of the ambient noise N is set in PD (0), the address and size of PD (0) are set in HDR (0), and HDR (0) Is added to the HDR queue of the sound output device 5 as HDR.

  Next, in the time interval t1, the audio data GD (1) obtained by adjusting the gain of the second divided audio data D (1) of the audio data D according to the volume level of the ambient noise N at that time is PD (1). And the address and size of PD (1) are set to HDR (1), and HDR (1) is added to the HDR queue of the sound output device 5 as HDR. Further, as indicated by Q {PD (0)}, the sound represented by GD (0) set in PD (0) is reproduced and output by the sound driver (output) 11 and the sound output device 5. However, in this example, when j is 1, this time after the time length ts has elapsed since the execution of the gain adjustment (step 214) for the divided audio data D (0) performed previously is started. This is for the case where the execution of the gain adjustment (step 214) of the divided audio data D (1) is started.

  Then, in the next time interval t2, the audio data GD (2) obtained by adjusting the gain of the third divided audio data D (2) of the audio data D according to the volume level of the ambient noise N at that time is PD. It is set to (0), the address and size of PD (0) are set to HDR (0), and HDR (0) is added to the HDR queue of the sound output device 5 as HDR. Further, as indicated by Q {PD (1)}, the sound represented by GD (1) set in PD (1) is reproduced and output by the sound driver (output) 11 and the sound output device 5.

  In the next time interval t3, the audio data GD (3) obtained by adjusting the gain of the fourth divided audio data D (3) of the audio data D according to the volume level of the ambient noise N at that time is PD ( 1), the address and size of the PD (1) are set to HDR (1), and HDR (1) is added to the HDR queue of the sound output device 5 as HDR. Further, as indicated by Q {PD (0)}, the sound represented by GD (2) set to PD (0) is reproduced and output by the sound driver (output) 11 and the sound output device 5.

  Thereafter, similarly, PD (0) and HDR (1) and HDR (0) and PD (1) are alternately used, and in the time interval tm, m + 1th divided audio data D (m ) And the sound output by the sound driver (output) 11 and the sound output device 5 represented by the sound data GD (m-1) after gain adjustment are performed.

As a result, as shown by Q, the sound obtained by adjusting the gain of the sound data D is output from the speaker 6 by the sound output device 5. Then, each part of the output voice is gain-adjusted according to the volume level of the ambient noise about ts time ago.
The first embodiment of the present invention has been described above.
As described above, according to the first embodiment, the audio data (PD) itself supplied to the sound driver (output) 11 by the audio output application 2 is subjected to gain adjustment according to the ambient noise. Regardless of the sound driver (output) 11, the gain adjustment of the output sound according to the ambient noise can be realized. In general, the interface API between the sound driver (output) 11 and the sound driver (input) 12 and the application is shared regardless of the sound driver (output) 11 and the sound driver (input) 12.

  Here, when the audio data (PD) itself supplied to the sound driver (output) 11 is subjected to gain adjustment according to ambient noise, the audio data (PD) is subjected to gain adjustment. A time difference between the time point and the time point when the sound represented by the sound data is actually output becomes a problem. This is because the output voice is gain-adjusted according to ambient noise in the past of the time difference.

  However, according to the first embodiment, the divided sound obtained by dividing the sound data representing the sound to be output by adjusting the gain of the sound data (PD) and supplying the sound data (PD) to the sound driver (output) 11. Since it can be performed in units of data, such a time difference can be shortened, and a sound that has been gain-adjusted according to the current ambient noise situation can be output. In addition, since it is considered that the ambient noise in the adjacent time is approximated, the sound subjected to gain adjustment in this way is practically presently used by sufficiently shortening the time length of the sound represented by the divided sound data. Can be used as a sound that has been gain-adjusted according to the ambient noise situation.

Hereinafter, a second embodiment of the present invention will be described.
In the second embodiment, the gain adjustment performed by the gain adjustment unit 22 can be executed more promptly in response to an instruction from the output control unit 21 in step 214 of the sound output processing of the first embodiment. is there.
FIG. 4 shows the configuration of the audio output device according to the second embodiment.
As illustrated, the speech element output device according to the second embodiment includes a noise data buffer 9 added to the speech output device illustrated in FIG. 1 and a noise data acquisition control unit 23 provided in the speech output application. Is.
In the second embodiment, in step 204 of the output control process performed by the output control unit 21 shown in FIG. 2, the calculated audio data division number n and the start of the audio output process are set to the noise data acquisition control unit 23 and the gain. Notify the adjustment unit.

The noise data acquisition control unit 23 performs the noise data acquisition process shown in FIG. 5a, and the gain adjustment unit 22 performs the output in step 214 of the audio output process of the first embodiment by the gain adjustment process shown in FIG. 5b. The gain is adjusted in accordance with an instruction from the control unit 21.
Hereinafter, the noise data acquisition process performed by the noise data acquisition control unit 23 will be described.
As shown in FIG. 5a, in this process, the sound data of the ambient noise is continuously captured from the sound driver (input) 12 until the output control unit 21 notifies the start of the sound output process (step 504), and the noise data buffer 9 NDPR is always stored with sound data of ambient noise for the latest past ts time (step 502). For example, NDPR in the noise data buffer 9 is always stored as a FIFO for storing audio data for ts time, and the sound data from the sound driver (input) 12 is sequentially stored. This can be realized, for example, by storing the captured ambient noise voice data in the NDPR. Note that ts is the time length of the divided audio data D (j) shown in the first embodiment.

  If the start of the audio output process is notified from the output control unit 21 (step 504), the sound driver is alternately switched to ND (0) and ND (1) of the noise data buffer 9 every ts time. (Input) The process of storing the ambient noise voice data fetched from 12 is performed n1 times (steps 506-514), and if stored n1 times, the process returns to step 502.

Next, gain adjustment processing performed by the gain adjustment unit 22 will be described.
As shown in FIG. 5b, in the gain adjustment processing, the output control unit 21 waits for a notification of the start of the audio output processing (step 552). If there is a notification, first, the divided audio data D (0) (step 554, 556), the gain adjustment is performed according to the volume level of the sound data of the ambient noise taken from the NDPR of the noise data buffer 9 (steps 558 and 572), and the sound data subjected to the gain adjustment is stored in the output buffer memory 8 as a PD. Store as (0) (steps 554, 566).

Thereafter, each j from j = 1 to j = n−1 is sequentially performed (steps 568 and 570).
The divided voice data D (j) is acquired (step 556), and the voice data of ambient noise for the time length ts is stored in ND (k) of the noise data buffer 9 (step 562). k) audio data of ambient noise is fetched from step (564), and the gain adjustment is performed on the obtained divided audio data D (j) according to the volume level of the acquired ambient noise audio data, and the gain data is adjusted. Is stored as PD (i) in the output buffer memory 8 (step 566). However, i is a remainder whose modulus is 2 of j, and k is a remainder whose modulus is 2 of j−1 (step 560).
When the processing up to j = n−1 is completed (step 568), the processing returns to step 552.

Next, FIG. 6 shows an example of gain adjustment processing as described above.
Now, as shown in FIG. 6a, the audio data D represents audio for a time length of 6 × ts, and the audio data D is divided into divided audio data from D (0) to D (5). Shall be. Further, as shown in FIG. 6b, the ambient noise N in each time interval t0 to t6 of each time length ts is represented by N (t0) to N (t6). However, the time interval t0 represents a time interval of the last past time length ts of the time point S at which the audio output process is started.

  In this case, when the audio output process is started at time S, at the time, the noise data acquisition control unit 23 stores the NDPR of the noise data buffer 9 in the most recent time interval t0 that already has the time length ts. Voice data N (t0) of ambient noise is stored. Therefore, the gain adjusting unit 22 immediately adjusts the gain according to the volume level of the ambient noise N (t0) stored in the NDPR to the first divided audio data D (0) of the audio data D, and the audio data GD. (0) is generated and set in PD (0) of the output audio buffer memory 8.

  Then, as soon as the sound data N (t1) of the ambient noise in the time interval t1 is stored in ND (0) by the noise data acquisition control unit 23, the gain adjustment unit 23nd the second divided sound of the sound data D. The audio data GD (1) is generated by adjusting the gain of the data D (1) according to the volume level of the ambient noise N (t1) stored in ND (0), and the PD of the output audio buffer memory 8 Set to (1).

  Next, the gain adjustment unit, as soon as the noise data acquisition control unit 23 stores the sound data N (t2) of the ambient noise in the time interval t2 in ND (1), the third divided sound of the sound data D The data D (2) is gain-adjusted according to the volume level of the ambient noise N (t2) stored in ND (1) to generate audio data GD (2), and the output audio buffer memory 8 PD Set to (0).

  Next, as soon as the noise data acquisition control unit 23 stores the sound data N (t3) of the ambient noise in the time interval t3 in the ND (0), the gain adjustment unit 4th divided sound data D of the sound data D (3) is gain-adjusted according to the volume level of the ambient noise N (t3) stored in ND (0) to generate audio data GD (3), and PD (1) of the output audio buffer memory 8 ).

Thereafter, similarly, while using the sound data of ambient noise stored in ND (1) and ND (0) alternately, gain adjustment is performed on the divided sound data D (4) to D (5), and PD (0) and PD (1) are set alternately.
As described above, according to the second embodiment, when the voice output request is generated, the voice data gain adjustment can be started promptly using the voice data of the latest ambient noise that is always held. It is also possible to start voice output promptly.

It is a block diagram which shows the structure of the audio | voice output apparatus which concerns on 1st Embodiment of this invention. It is a flowchart which shows the audio | voice output process which concerns on 1st Embodiment of this invention. It is a figure which shows the process example of the audio | voice output process which concerns on 1st Embodiment of this invention. It is a block diagram which shows the structure of the audio | voice output apparatus which concerns on 2nd Embodiment of this invention. It is a flowchart which shows the noise data acquisition process and gain adjustment process which concern on 2nd Embodiment of this invention. It is a figure which shows the process example of the noise data acquisition process and gain adjustment process which concern on 2nd Embodiment of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Operating system, 2 ... Sound output application, 3 ... Sound input device, 4 ... Microphone, 5 ... Sound output device, 6 ... Speaker, 7 ... Sound data memory, 8 ... Output buffer memory, 9 ... Noise data buffer, DESCRIPTION OF SYMBOLS 11 ... Sound driver (output), 12 ... Sound driver (input), 21 ... Output control part, 22 ... Gain adjustment part, 23 ... Noise data acquisition control part

Claims (6)

A microphone,
A sound input device for generating ambient audio data representing ambient audio picked up using the microphone;
Memory storing original audio data,
A buffer for storing audio data;
A sound output device for outputting sound to a speaker ;
The issued voice output request is accepted, the voice data stored in the buffer specified as the output voice data in the accepted voice output request is read, and the voice represented by the read voice data is output to the speaker. An output audio gain adjustment device applied to an audio output system including a sound driver to be performed by the sound output device ,
Noise level detection means for detecting a volume level of ambient noise represented by ambient audio data generated by the sound input device;
When the output of the original sound data stored in the memory is requested, the original sound data stored in the memory is divided into divided sound data that is sound data representing sound for each predetermined time length. The divided sound data is divided into target divided sound data sequentially at the predetermined time intervals, and the sound represented by the target divided sound data is volume of ambient noise detected by the noise level detecting means at the time Generating the audio data representing the audio amplified by a gain according to a level, storing the generated audio data in the buffer, and outputting the audio output request specifying the stored audio data as the output audio data Output audio gain adjustment comprising audio output processing means for performing audio output processing issued to a sound driver Location. The gain adjusting device according to claim 1,
The noise level detection means includes
Ambient noise holding means for holding the surrounding voice data for the predetermined time length in the past past, generated by the sound input device, at least until the start of the voice output processing of the voice output processing means,
At the start of the sound output processing of the sound output processing means, the ambient noise to be detected is detected at the time when the volume level of the surrounding sound data for the predetermined time length in the latest past held in the ambient noise holding means Thereafter, the volume level of the surrounding audio data generated by the sound input device during the most recently passed time interval for each elapse of the time interval of the predetermined time length, Volume level calculating means for calculating the volume level of the ambient noise to be detected;
The sound output processing means, in the sound output processing, each time the sound volume level calculating means calculates the sound volume level of the ambient noise, the divided sound data is sequentially set as the target divided sound data, and the target An audio output device that generates the audio data representing the audio obtained by amplifying the audio represented by the divided audio data with a gain corresponding to the calculated volume level of the ambient noise. The output audio gain adjusting device according to claim 1 or 2 ,
The noise level detection means detects the volume level of ambient noise for each frequency band,
The sound output processing means amplifies the sound represented by the target divided sound data with a gain corresponding to the volume level of each frequency band of the ambient noise detected by the noise level detection means at each time point for each frequency band. An output audio gain adjustment apparatus, characterized in that the audio data representing the reproduced audio is generated. The output audio gain adjusting device according to claim 1, 2, or 3 ,
The audio output device is an audio output gain adjustment device mounted on an automobile. A microphone,
A sound input device for generating ambient audio data representing ambient audio picked up using the microphone;
Memory storing original audio data,
A buffer for storing audio data;
A sound output device for outputting sound to a speaker ;
The issued voice output request is accepted, the voice data stored in the buffer specified as the output voice data in the accepted voice output request is read, and the voice represented by the read voice data is output to the speaker. A computer program that is read and executed by a computer having a sound driver for the sound output device ,
The computer,
Noise level detection means for detecting a volume level of ambient noise represented by ambient audio data generated by the sound input device;
When the output of the original sound data stored in the memory is requested, the original sound data stored in the memory is divided into divided sound data that is sound data representing sound for each predetermined time length. The divided sound data is divided into target divided sound data sequentially at the predetermined time intervals, and the sound represented by the target divided sound data is volume of ambient noise detected by the noise level detecting means at the time Generating the audio data representing the audio amplified by a gain according to a level, storing the generated audio data in the buffer, and outputting the audio output request specifying the stored audio data as the output audio data A computer characterized by functioning as an audio output processing means for performing an audio output process issued to a sound driver Program. A computer program according to claim 5,
The noise level detection means includes
Ambient noise holding means for holding the surrounding voice data for the predetermined time length in the past past, generated by the sound input device, at least until the start of the voice output processing of the voice output processing means,
At the start of the sound output processing of the sound output processing means, the ambient noise to be detected is detected at the time when the volume level of the surrounding sound data for the predetermined time length in the latest past held in the ambient noise holding means Thereafter, the volume level of the surrounding audio data generated by the sound input device during the most recently passed time interval for each elapse of the time interval of the predetermined time length, Volume level calculating means for calculating the volume level of the ambient noise to be detected;
The sound output processing means, in the sound output processing, each time the sound volume level calculating means calculates the sound volume level of the ambient noise, the divided sound data is sequentially set as the target divided sound data, and the target A computer program for generating sound data representing sound obtained by amplifying sound represented by divided sound data with a gain corresponding to the volume level of the calculated ambient noise