JP2018125822A - In-vehicle conversation support device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an in-vehicle conversation support device capable of improving difficulty in hearing an utterance content occurring in accordance with an utterance volume.SOLUTION: An in-vehicle conversation support device 1 comprises: a microphone 10 that collects a voice of a passenger in a front row; a microphone 30 that collects the voice of the passenger in a rear row; a voice processing part 12 that outputs the voice collected by the microphone 10 from a speaker 36 toward the passenger in the rear row; an amplifier 14; a voice processing part 32 that outputs the voice collected by the microphone 30 from a speaker 16 toward the passenger in the front row; an amplifier 34; a voice output stop determination unit 18 that stops the output of the voice toward the passenger of the rear row from the speaker 36 when an utterance voice of the passenger in the front row reaches to the passenger in the rear row; and a voice output stop determination part 38 that stops the output of the voice toward the passenger in the front row from the speaker 16 when achieving the utterance voice of the passenger in the rear row to the passenger in the front row.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an in-car conversation support apparatus that is mounted on a vehicle and that collects a utterance voice of a passenger with a microphone and outputs it from a speaker.

  Conventionally, the voices of passengers in the driver's seat and passenger's seat are collected with a microphone, or the voices of passengers in the backseat are collected with a microphone and output from a speaker. 2. Description of the Related Art A vehicle interior conversation assist device that makes it easy for other passengers to hear the content of a passenger's utterance is known (see, for example, Patent Document 1).

  Note that an in-car conversation support apparatus that performs voice recognition processing on a passenger's uttered voice and displays the utterance content on a display as a text has been conventionally known (see, for example, Patent Document 2).

JP 2010-83452 A JP 2014-170154 A

  By the way, in the in-vehicle conversation assistance device disclosed in Patent Document 1 described above, when the volume of the occupant speaking is high, the output sound that is output from the speaker and the uttered sound itself are overlapped with another passenger. There was a problem that it was difficult to hear the utterance content. For example, if the utterance volume of the passenger in the rear seat is high, this utterance voice reaches the passenger in the driver's seat or passenger seat directly, and the same sound is output from the speaker after a delay caused by voice processing etc. As a result, it reaches the driver's seat and passenger's seat passenger, which is annoying for passengers who want to understand the contents of these voices.

  In addition, when the utterance content is displayed in text as in the in-car conversation support device disclosed in Patent Document 2 described above, the arrival timing is shifted from each other and the voice does not reach, but other passengers Because it is necessary to read the displayed text to understand the content uttered at a volume that cannot be directly understood, it increases the burden of conversation and lacks the realism of the conversation. I can not say.

  The present invention was created in view of the above points, and an object of the present invention is to provide an in-vehicle conversation support device that can improve difficulty in hearing the utterance content generated according to the utterance volume. .

  In order to solve the above-described problem, the in-vehicle conversation support device of the present invention includes a first microphone that collects the voice of the passenger in the front row, a second microphone that collects the voice of the passenger in the rear row, First sound output means for outputting the sound collected by the first microphone from the speaker toward the passenger in the rear row, and outputting the sound collected by the second microphone from the speaker toward the passenger in the front row. The second voice output means and the first voice that stops outputting the voice toward the passenger in the rear row by the first voice output means when the voice of the passenger in the front row reaches the passenger in the rear row Output canceling means, and second voice output stopping means for canceling the output of the voice toward the front row passengers by the second voice output means when the voice of the rear row passengers reaches the front row passengers And.

  Audio output using the speaker is canceled when the voice of the passenger in the rear row reaches the passenger in the front row directly or when the voice of the passenger in the front row reaches the passenger in the rear row directly Therefore, it is possible to prevent the voices of the same utterance content from reaching at a timing shifted, and it is possible to improve the difficulty in hearing when the utterance volume is high.

  Further, when the presence or absence of the utterance voice of the front row passengers reaching the rear row passengers is repeatedly switched, the contents of the front row passengers' voices collected by the first microphone are analyzed and the back row passenger speech is analyzed. Boarding in the back row collected by the second microphone when the first speech recognition processing means to be displayed to the passenger and the presence / absence of the speech of the back row passengers reaching the front row passenger are repeatedly switched It is desirable to further include second voice recognition processing means for analyzing the content of the person's uttered voice and displaying it to the front row passengers. As a result, the utterance voice of the passengers in the rear row has reached the passengers in the front row directly, but the volume is insufficient and the contents can only be heard partially, or If you have reached the passenger in the back row directly, but the volume is insufficient and you can only hear the content, you can check the content of the utterance voice on the display, making it difficult to hear due to insufficient volume. Can be complemented.

  In addition, the first voice output stopping unit described above is configured so that the first voice recognition processing unit displays the first voice when the analysis result of the utterance voice of the passenger in the front row is displayed toward the passenger in the rear row. The output means stops outputting the voice toward the passenger in the back row, and the second voice output stop means outputs the analysis result of the utterance voice of the passenger in the back row to the passenger in the front row by the second voice recognition processing means. It is desirable to stop outputting the sound directed to the passengers in the front row by the second sound output means when the information is being displayed. As a result, the utterance voice of the passengers in the rear row has reached the passengers in the front row directly, but the volume is insufficient and the contents can only be heard partially, or Directly reaching the passengers in the rear row, but when the volume is insufficient and the contents can only be heard partially, the harsh sound that is generated in fragments is made by preventing the sound from being output from the speaker. It becomes possible to avoid duplication.

  In addition, the first audio output stopping unit described above determines whether or not the utterance voice of the front row passenger has reached the back row passenger based on the audio signal output from the second microphone, Preferably, the second audio output canceling means determines whether or not the utterance voice of the rear row passenger has reached the front row passenger based on the audio signal output from the first microphone. The second microphone that collects the voice of the passenger in the rear row is used to detect the voice of the passenger in the front row, and the first microphone that collects the voice of the passenger in the front row is used to detect the voice of the passenger in the rear row. By detecting the sound, it is not necessary to prepare a separate structure for detecting the sound, so that the cost can be reduced by reducing the number of parts.

  Further, the third microphone that collects the voice of the passenger in the rear row in the vicinity of the listening position of the passenger in the front row and the fourth microphone that collects the voice of the passenger in the front row in the vicinity of the listening position of the passenger in the rear row. The first audio output stopping means determines whether or not the utterance voice of the front row passenger has reached the rear row passenger based on the audio signal output from the fourth microphone. Then, it is desirable that the second audio output stopping means determines whether or not the utterance voice of the passenger in the rear row has reached the passenger in the front row based on the audio signal output from the third microphone. As a result, it is possible to more accurately determine whether or not the voices of other passengers reach each passenger directly, and to appropriately switch the presence or absence of voice output from the speaker in order to improve difficulty in hearing. It becomes possible.

It is a figure which shows the structure of the in-vehicle conversation assistance apparatus of 1st Embodiment. It is a figure which shows the relationship between the volume of the passenger's speech sound detected with a microphone, and the presence or absence of the sound output from a speaker. It is a figure which shows the structure of the in-vehicle conversation assistance apparatus of 2nd Embodiment. It is a figure which shows the relationship between the volume of the passenger's speech sound detected with a microphone, the presence or absence of the audio | voice output from a speaker, and the presence or absence of the audio | voice content display by a display part. It is a figure of the modification which shows the relationship between the volume of the passenger's speech sound detected with a microphone, the presence or absence of the audio | voice output from a speaker, and the presence or absence of the audio | voice content display by a display part. It is a figure which shows the structure of the modification of the in-vehicle conversation assistance apparatus which added the microphone for audio | voice detection. It is a figure which shows the structure of the other modification of the in-vehicle conversation assistance apparatus which added the microphone for audio | voice detection.

  Hereinafter, an in-vehicle conversation support device according to an embodiment to which the present invention is applied will be described with reference to the drawings.

(First embodiment)
FIG. 1 is a diagram illustrating a configuration of the in-vehicle conversation support device according to the first embodiment. The in-car conversation support apparatus 1 shown in FIG. 1 includes microphones 10 and 30, audio processing units 12 and 32, amplifiers 14 and 34, speakers 16 and 36, and audio output stop determination units 18 and 38. This in-car conversation support device is mounted on a vehicle.

  The microphone 10 collects the voice of the passenger in the front row of the host vehicle. The microphone 10 is attached to a position where the voices of the driver and the passenger in the passenger seat can be collected, for example, the dashboard or the ceiling at the upper center of the front row seat.

  The microphone 30 collects the voice of the passenger behind the host vehicle. The microphone 30 is attached to a position where the voices of the passengers in the rear row can be collected, for example, a headrest of a driver seat or a passenger seat, a ceiling in the upper center of the rear row, and the like.

  The sound processing unit 12 removes components unrelated to the collected sound such as wind noise and road noise from the sound collected by the microphone 10 for the front row, and emphasizes the frequency component of the sound band. Process. The sound after the completion of these processes is amplified by the amplifier 14 and output from the rear-row speaker 36. The speakers 36 are mounted at positions suitable for listening to the passengers in the rear row, for example, on the left and right sides of the rear row seats.

  The sound processing unit 32 removes components unrelated to the collected sound such as wind noise and road noise from the sound collected by the microphone 30 for the back row, or emphasizes the frequency component of the sound band. Process. The sound after the completion of these processes is amplified by the amplifier 34 and output from the front-row speaker 16. The speakers 16 are mounted at positions suitable for listening to the passengers in the front row, for example, on the left and right sides of the driver seat and the passenger seat.

  The voice output stop determination unit 18 stops the output of the voice directed to the passengers in the rear row by the voice processing unit 12, the amplifier 14, and the speaker 36 when the voice of the passengers in the front row reaches the passengers in the rear row. . Specifically, the audio output cancellation determination unit 18 determines whether the utterance voice of the front row passenger has reached the rear row passenger based on the audio signal output from the microphone 30. The voices of the passengers in the front row are clearly collected by the front row microphone 10 and can be separated from ambient noise such as road noise. Further, when the volume of the uttered voice of the passengers in the front row is high, the sound should be collected by the rear row microphone 30 at approximately the same time. Therefore, by monitoring the presence / absence of uttered sound collected by the rear-stage microphone 30 at the timing of collecting the sound by the front-row microphone 10 (a slight delay occurs due to passing through the voice processing unit 12 or the like). It is possible to determine whether or not the utterance voice of the front row passenger has been collected by the rear row microphone 30, that is, whether or not the utterance voice of the front row passenger has reached the rear row passenger. .

  The voice output stop determination unit 38 stops the output of the voice toward the front row passengers by the voice processing unit 32, the amplifier 34, and the speaker 16 when the voice of the rear row passengers reaches the front row passengers. . Specifically, the audio output stop determination unit 38 determines whether the utterance voice of the passenger in the rear row has reached the passenger in the front row based on the audio signal output from the microphone 10. Voices of passengers in the rear row are clearly collected by the rear row microphone 30 and can be separated from ambient noise such as road noise. Further, when the volume of the uttered voice of the passengers in the rear row is high, the sound should be collected by the microphone 10 for the front row at approximately the same time. Therefore, by monitoring the presence / absence of the uttered sound collected by the front row microphone 10 at the timing of collecting the sound by the rear row microphone 30 (a slight delay occurs due to passing through the voice processing unit 32). Thus, it is possible to determine whether or not the utterance voice of the passenger in the rear row has been collected by the microphone 10 for the front row, that is, whether or not the voice of the passenger in the rear row has reached the passenger in the front row. .

  The audio processing unit 12 and the amplifier 14 described above are used as the first audio output unit, the audio processing unit 32 and the amplifier 34 are used as the second audio output unit, and the audio output stop determination unit 18 is used as the first audio output stop unit. The sound output stop determination unit 38 corresponds to the second sound output stop unit.

  FIG. 2 is a diagram showing the relationship between the volume of the uttered voice of the passenger detected by the microphones 10 and 30 and the presence or absence of the voice output from the speakers 16 and 36. FIG. 2A shows the relationship between the volume of the utterance voice of the front row passenger detected using the rear row microphone 30 and the presence or absence (ON / OFF) of the voice output from the speaker 36. B) shows the relationship between the volume of the uttered voice of the passenger in the rear row detected using the front row microphone 10 and the presence or absence (ON / OFF) of the voice output from the speaker 16.

  As shown in FIG. 2A, when the volume of the utterance voice of the front row passengers collected by the rear row microphone 30 is smaller than the reference value a, that is, the volume of the utterance voice of the front row passengers is When it is small and does not reach the passengers in the rear row directly, the system of the voice processing unit 12, the amplifier 14, and the speaker 36 is effective, and the voices of the passengers in the front row are collected and amplified and output from the speaker 36. (Speaker 36 ON). The passengers in the rear row can understand the contents of the uttered voices of the passengers in the front row by listening to the sound output from the speaker 36.

  However, when the volume of the utterance voice of the front row passengers collected by the rear row microphone 30 is larger than the reference value a, that is, the volume of the utterance voice of the front row passengers is large and directly to the rear row passengers. When reaching, the system of the voice processing unit 12, the amplifier 14, and the speaker 36 is invalidated, and the voice of the passenger in the front row is not output from the speaker 36 (speaker 36 OFF). However, since the volume of the utterance voice of the passengers in the front row is high, the passengers in the rear row can directly hear the utterance voices and can understand the contents of the utterance voices of the passengers in the front row. .

  Further, as shown in FIG. 2B, when the volume of the utterance voice of the rear row passengers collected by the front row microphone 10 is smaller than the reference value A, that is, the utterance voice of the rear row passengers. When the volume is low and does not reach the front row passengers directly, the system of the voice processing unit 32, the amplifier 34, and the speaker 16 is effective, and the voices of the passengers in the rear row are collected and amplified, and the speaker 16 (Speaker 16 ON). The passengers in the front row can understand the contents of the uttered voices of the passengers in the rear row by listening to the sound output from the speaker 16.

  However, if the volume of the utterance voice of the rear row passengers collected by the microphone 10 for the front row is larger than the reference value A, that is, the volume of the utterance voice of the rear row passengers is large and directly to the front row passengers. When reaching, the system of the audio processing unit 32, the amplifier 34, and the speaker 16 becomes invalid, and the utterance voice of the passenger in the rear row is not output from the speaker 16 (speaker 16 OFF). However, since the volume of the utterance voice of the passenger in the rear row is high, the passenger in the front row can directly hear the voice of the utterance and can understand the content of the voice of the passenger in the rear row. .

  Note that the reference values a and A for enabling / disabling audio output from the speakers 16 and 36 do not have to be the same value, and are set to different values depending on the microphones 10 and 30 and the arrangement state of each passenger. It is desirable to do.

  As described above, in the in-car conversation support device 1 according to the present embodiment, the utterance voice of the passenger in the rear row directly reaches the passenger in the front row, or the utterance voice of the passenger in the front row directly reaches the passenger in the rear row. In such a case, the voice output using the speakers 16 and 36 is stopped in such a case, so that the voice of the same utterance content is shifted in two systems (a system that is directly transmitted and a system that is transmitted via the speaker). Can be prevented, and it is possible to improve the difficulty in hearing when the speech volume is high.

  In addition, the microphone 30 that collects the voice of the passenger in the rear row is used to detect the voice of the passenger in the front row, and the voice of the passenger in the rear row is obtained using the microphone 10 that collects the voice of the passenger in the front row. By detecting, it is not necessary to prepare a separate configuration for voice detection, so that the cost can be reduced by reducing the number of parts.

(Second Embodiment)
FIG. 3 is a diagram illustrating a configuration of the in-vehicle conversation support device according to the second embodiment. 3 is different from the in-car conversation support device 1 shown in FIG. 1 in that voice recognition processing units 20 and 40 and display units 22 and 42 are added. In the following, the description will be given focusing on this added configuration.

  The voice recognition processing unit 20 determines whether the front row of passengers collected by the microphone 10 when the presence / absence of the speech of the front row passengers reaching the back row passengers is repeatedly switched (when it is heard or not heard). The content of the speech is analyzed and displayed for the passengers in the back row. Specifically, the content of the audio obtained by this analysis is displayed on the display unit 42 installed at a position where the passenger in the rear seat can see. The displayed audio content may be hiragana display, but it is desirable to display kanji mixed with kana.

  The voice recognition processing unit 40, when the presence or absence of the utterance voice of the passengers in the rear row reaches the front row passengers is repeatedly switched (when it can be heard or not heard), Analyzes the content of the speech and displays it to the passengers in the front row. Specifically, the content of the voice obtained by this analysis is displayed on the display unit 22 installed at a position where the driver and the passenger in the passenger seat can see. The displayed audio content may be hiragana display, but it is desirable to display kanji mixed with kana.

  The voice recognition processing unit 20 described above corresponds to the first voice recognition processing unit, and the voice recognition processing unit 40 corresponds to the second voice recognition processing unit.

  FIG. 4 is a diagram showing the relationship between the volume of the uttered voice of the passenger detected by the microphones 10 and 30, the presence or absence of voice output from the speakers 16 and 36, and the presence or absence of voice content display by the display units 22 and 42. is there. FIG. 4A shows the volume of the utterance voice of the front row passenger detected using the rear row microphone 30, the presence / absence (ON / OFF) of the voice output from the speaker 36, and the voice content display by the display unit 42. The relationship with the presence / absence (ON / OFF) is shown. 4B shows the volume of the utterance voice of the passengers in the rear row detected by using the front row microphone 10, the presence or absence (ON / OFF) of the voice output from the speaker 16, and the voice from the display unit 22. The relationship with the presence / absence of content display (ON / OFF) is shown.

  As shown in FIG. 4A, when the utterance voice of the front row passengers collected by the rear row microphone 30 is smaller than the reference value b1, that is, the volume of the utterance voice of the front row passengers is low. When the passengers in the rear row are not reached at all, the system of the voice processing unit 12, the amplifier 14, and the speaker 36 is effective, and the voices of the passengers in the front row are collected and amplified and output from the speaker 36. (Speaker 36 ON). The passengers in the rear row can understand the contents of the uttered voices of the passengers in the front row by listening to the sound output from the speaker 36. Further, when the volume of the utterance voice of the front row passengers collected by the rear row microphone 30 is higher than the reference value b2 (> b1), that is, the volume of the utterance voice of the front row passengers is large and When the passenger directly reaches the passenger, the system of the voice processing unit 12, the amplifier 14, and the speaker 36 is disabled, and the speech voice of the passenger in the front row is not output from the speaker 36 (speaker 36 OFF). However, since the volume of the utterance voice of the passengers in the front row is high, the passengers in the back row can directly hear this utterance voice and can clearly understand the contents of the utterance voices of the passengers in the front row It becomes. In these cases, since the content of the utterance voice can be understood from the output voice of the speaker 36 or the voice that directly reaches, the analysis result of the utterance voice of the passenger in the front row by the voice recognition processing unit 20 is It is not displayed on the display unit 42 (display OFF).

  By the way, when the volume of the utterance voice of the front row passengers collected by the rear row microphone 30 is between the reference values b1 and b2, that is, the volume of the utterance voice of the front row passengers is so small that it cannot be heard at all. Is not large enough to be heard clearly, and if the presence or absence of the utterances of the passengers in the front row reaches the passengers in the back row is repeatedly switched and can be heard or cannot be heard, the voice recognition processing unit 20 The analysis result of the uttered voice of the passenger is displayed on the display unit 42 in the rear row. In this case, the system of the audio processing unit 12, the amplifier 14, and the speaker 36 is effective, and the voices of the passengers in the front row are collected and amplified and output from the speaker 36 (speaker 36 ON). .

  Also, as shown in FIG. 4B, when the utterance voice of the rear row passengers collected by the front row microphone 10 is smaller than the reference value B1, that is, the volume of the utterance voice of the rear row passengers is When it is small and does not reach the front row passengers at all, the system of the voice processing unit 32, the amplifier 34, and the speaker 16 is effective, and the voices of the back row passengers are collected and amplified and output from the speaker 16 (Speaker 16 ON). The passengers in the front row can understand the contents of the uttered voices of the passengers in the rear row by listening to the sound output from the speaker 16. Further, when the volume of the utterance voice of the rear row passengers collected by the front row microphone 10 is larger than the reference value B2 (> B1), that is, the volume of the utterance voice of the rear row passengers is large, When directly reaching the occupant, the system of the audio processing unit 32, the amplifier 34, and the speaker 16 is disabled, and the utterance voice of the occupant in the rear row is not output from the speaker 16 (speaker 16 OFF). However, since the voices of the passengers in the back row are loud, the passengers in the front row can directly hear the voices of the passengers in the back row and can clearly understand the contents of the voices of the passengers in the back row. It becomes. In these cases, since the content of the uttered voice can be understood by the output voice of the speaker 16 or the voice that directly reaches, the analysis result of the uttered voice of the passenger in the rear row by the voice recognition processing unit 40 is It is not displayed on the display unit 22 (display OFF).

  By the way, when the volume of the uttered voice of the rear row passengers collected by the front row microphone 10 is between the reference values B1 and B2, that is, the volume of the uttered voice of the rear row passengers is so small that it cannot be heard at all. Is not large enough to be heard clearly, and if the back row passenger's utterance voice has reached the front row passengers repeatedly, it can be heard or cannot be heard. The analysis result of the uttered voice of the passenger is displayed on the display unit 22 in the front row. In this case, the system of the audio processing unit 32, the amplifier 34, and the speaker 16 is effective, and the uttered voice of the passenger in the rear row is collected and amplified and output from the speaker 16 (speaker 16 ON). .

  As described above, in the in-car conversation support apparatus 1A of the present embodiment, the uttered voice of the passenger in the rear row reaches the passenger in the front row directly, but the volume is insufficient so that the contents can be heard only partially. If the utterance voice of the passenger in the front row reaches the passenger in the rear row directly, but the volume is insufficient and the content can only be heard partially, the content of the utterance voice can be displayed. This makes it possible to confirm the difficulty of hearing due to insufficient volume.

  By the way, in the second embodiment described above, as shown in FIG. 4, while the analysis result of the spoken voice is displayed (display ON), the voice output is performed from the speakers 16 and 36 ( (Speaker 16 ON, speaker 36 ON) As shown in FIG. 5, while the analysis result of the uttered voice is displayed (display ON), the voice output from the speakers 16 and 36 may be stopped ( Speaker 16 OFF, speaker 36 OFF). This prevents audio with the same utterance content from being partially misaligned in two systems (systems that are directly transmitted and systems that are transmitted via speakers), and prevents duplication of harsh audio that occurs in fragments. It can be avoided.

  In addition, this invention is not limited to the said embodiment, A various deformation | transformation implementation is possible within the range of the summary of this invention. For example, in the above-described embodiment, it is detected by using the rear row microphone 30 that the utterance voice of the front row passenger reaches the rear row passenger, and the rear row passenger speech sound reaches the front row passenger. This is detected using the front row microphone 10, but another microphone may be used for detecting these sounds.

  FIG. 6 is a diagram showing a configuration of a modified example of the in-vehicle conversation support apparatus in which a microphone is added for voice detection. The configuration shown in FIG. 6 is different from the configuration shown in FIG. 1 in that microphones 24 and 44 are added.

  FIG. 7 is a diagram showing a configuration of another modification of the in-vehicle conversation support apparatus to which a microphone is added for voice detection. The configuration shown in FIG. 7 is different from the configuration shown in FIG. 3 in that microphones 24 and 44 are added. Hereinafter, the description will be given focusing on the microphones 24 and 44 added in these modified examples.

  One microphone 24 is for collecting the voices of the passengers in the rear row near the listening position of the passengers in the front row, and is installed, for example, in the headrest of the driver's seat corresponding to the position of the driver's ear. ing. The other microphone 44 is for collecting the voice of the front row passengers in the vicinity of the listening position of the rear row passengers. For example, the rear row passengers (one of the two left and right passengers) It is installed in the headrest of the rear seat corresponding to the position of the ears. In addition, these installation positions are examples, and may be changed as appropriate.

  The audio output cancellation determination unit 18 determines whether or not the utterance voice of the passenger in the front row has reached the passenger in the rear row based on the audio signal output from the microphone 44 installed in the rear seat. The voice output stop determination unit 38 determines whether or not the utterance voice of the passenger in the rear row has reached the passenger in the front row based on the voice signal output from the microphone 24 installed in the driver's seat.

  In this way, by using the microphones 24 and 44 added to detect the presence or absence of voice arrival, it is possible to more accurately determine whether or not the voices of other passengers directly reach each passenger. Therefore, it is possible to appropriately switch the presence or absence of audio output from the speakers 16 and 36 in order to improve difficulty in hearing.

  Further, in the above-described embodiment, the present invention is applied to a two-row seat vehicle including a front row and a rear row, but the present invention can also be applied to a vehicle having three or more row seats.

  As described above, according to the present invention, the utterance voice of the passenger in the rear row directly reaches the passenger in the front row, or the utterance voice of the passenger in the front row directly reaches the passenger in the rear row. In this case, since the sound output using the speaker is stopped, it is possible to prevent the voice of the same utterance content from reaching when it is shifted, and it is possible to improve the difficulty in hearing when the utterance volume is high It becomes.

1 In-car conversation support device 10, 24, 30, 44 Microphone 12, 32 Audio processing unit 14, 34 Amplifier 16, 36 Speaker 18, 38 Audio output stop determination unit 20, 40 Voice recognition processing unit 22, 42 Display unit

Claims (5)

A first microphone that collects the voices of the front row passengers;
A second microphone that collects the voices of the passengers in the back row;
First sound output means for outputting sound collected by the first microphone from a speaker toward a passenger in the rear row;
Second sound output means for outputting the sound collected by the second microphone from the speaker toward the front row passengers;
First voice output stopping means for stopping the output of the voice directed to the passengers in the rear row by the first voice output means when the voice of the passenger in the front row reaches the passengers in the rear row; ,
Second audio output stopping means for stopping the output of the voice toward the front row passengers by the second audio output means when the uttered voice of the passengers in the rear row reaches the passengers in the front row; ,
An in-car conversation support device comprising: When the presence / absence of the speech of the front row passengers reaching the back row passengers is repeatedly switched, the back row is analyzed by analyzing the content of speech signals of the front row passengers collected by the first microphone. First speech recognition processing means for displaying to a passenger of the vehicle,
When the presence or absence of the utterance voice of the passengers in the rear row is repeatedly switched, the contents of the voices of the passengers in the rear row collected by the second microphone are analyzed and the front row is analyzed. Second voice recognition processing means for displaying to a passenger of the vehicle,
The in-vehicle conversation support device according to claim 1, further comprising: The first voice output stopping unit is configured to display the first voice recognition processing unit when the analysis result of the utterance voice of the passenger in the front row is displayed toward the passenger in the rear row. Stop outputting the voice to the passengers in the back row by the voice output means,
The second voice output stopping unit is configured to display the second voice recognition processing unit when the analysis result of the utterance voice of the passenger in the rear row is displayed toward the passenger in the front row. The in-vehicle conversation support device according to claim 2, wherein the voice output means stops outputting the voice toward the front row passengers. The first audio output stopping means determines whether or not the utterance voice of the front row passenger has reached the rear row passenger based on the audio signal output from the second microphone,
The second audio output stopping means determines whether or not the utterance voice of the passenger in the rear row has reached the passenger in the front row, based on the audio signal output from the first microphone. The in-vehicle conversation support device according to any one of claims 1 to 3. A third microphone for collecting the sound of the passengers in the rear row near the listening position of the passengers in the front row;
A fourth microphone for collecting the sound of the passengers in the front row near the listening position of the passengers in the rear row;
The first audio output stop means further determines whether the utterance voice of the front row passenger has reached the rear row passenger based on the audio signal output from the fourth microphone. Judgment,
The second audio output stopping means determines whether or not the utterance voice of the rear row passenger has reached the front row passenger based on an audio signal output from the third microphone. The in-vehicle conversation support device according to any one of claims 1 to 3.

Images