JP4652709B2 - On-vehicle acoustic device, control method, and program - Google Patents

Description

  The present invention relates to an on-vehicle acoustic device, a control method, and a program that take into account passenger conversation.

  Patent Document 1 discloses that a passenger's conversation and noise are detected separately, and the volume of the acoustic device is automatically lowered during a conversation period and automatically raised during a non-conversation period.

  In Patent Document 2, when the volume of the route guidance of the car navigation device is automatically increased according to the level of noise, the noise in the passenger compartment is detected by a microphone. Since the conversation sound is detected as noise along with noise such as wind noise and audio playback sound, as a result, the volume of the route guidance automatically increases even during the conversation period, and the conversation is disturbed. It is disclosed that noise is detected from a motor current of a blower of an air conditioner, a window opening / closing switch or the like by eliminating a microphone.

Patent Document 3 discloses a technique for localizing a virtual sound image in front of the passenger and at the eye level of the passenger, while the left and right speakers are disposed below the left and right doors.
JP 2002-171588 A JP-A-5-332778 JP 2002-95096 A

  It is desirable that the reproduced sound from the speaker does not disturb the conversation. In the in-vehicle acoustic devices disclosed in Patent Documents 1 and 2, the fact that the conversation is interrupted is achieved by automatically decreasing or not increasing the volume from the speaker.

  Patent Document 3 discloses generating a virtual sound image at a predetermined height, but does not mention any control of the virtual sound image during the conversation period.

  An object of the present invention is to provide an in-vehicle acoustic device, a control method, and a program that prevent conversation disturbance due to reproduced sound during conversation without adjusting the volume.

The on-vehicle acoustic device of the present invention includes a plurality of speakers arranged at different positions in the vehicle interior. The in-vehicle acoustic device has the following means.
The conversation period detecting means for detecting whether the period is a conversation period or a non-conversation period based on the utterance of the passenger, and a virtual sound image generated by sound emission from a plurality of speakers is almost A correction signal is generated by correcting the reproduction signal from the media so that the virtual sound image disappears from the height of the ear of the passenger so that the virtual sound image disappears during the conversation period. Reproduction signal correction means for outputting a correction signal to each speaker.

The on-vehicle acoustic device to which the control method of the present invention is applied includes a plurality of speakers arranged at different positions in the vehicle interior, and the control method includes the following steps.
A conversation period detection step for detecting whether the conversation period or non-conversation period based on the utterance of the passenger,
The virtual sound image generated by sound emission from a plurality of speakers is almost the height of the passenger during the non-conversation period, and the virtual sound image is almost the ear of the passenger during the conversation period. A reproduction signal correction step of generating a correction signal obtained by correcting the reproduction signal from the media so as to disappear from the height of the medium and outputting the correction signal to each speaker.

  The program of the present invention is a program for causing a computer to function as each means of the above-described on-vehicle acoustic device.

  According to the present invention, during the conversation period, the virtual sound image at almost the ear height of the passenger disappears, so that the passenger is prevented from disturbing the conversation due to the reproduced sound from the speaker.

  1 and 2 are a schematic side view and a schematic plan view of the automobile 10 showing an arrangement example of the speakers 18a and 18b in the passenger compartment 11. FIG. In the passenger compartment 11, the driver's seat 12 and the passenger seat 13 are disposed on the front side, and the rear seat 14 is disposed on the rear side. In the illustrated example, the passenger 16 is seated in the driver's seat 12 and the passenger seat 13 respectively. The passenger 16 may be present in the rear seat 14. The left and right speakers 18a, 18b are arranged on the inner surface side of the left and right front doors and in the lower half. The position of the lower half is a position near the knee of the seated passenger 16.

  FIG. 3 is a hardware configuration diagram of the in-vehicle acoustic device 25. CD28, MD29, and DVD30 as media and sources are played back audio by a corresponding player, and the playback signal is sent to selector 34. The tuner 31 receives a radio broadcast as a broadcast medium and outputs the demodulated audio signal as a reproduction signal to the selector 34. The selector 34 selects any one of the reproduction signals of the CD 28, the MD 29, the DVD 30 and the tuner 31, and sends the reproduction signal related to the selected device to the sound field control unit 37 and the conversation detection circuit 39. The conversation detection circuit 39 receives the reproduction signal from the selector 34 together with the output signal of the microphone 40. The microphone 40 is adjusted in directivity toward the face of the passenger 16 and detects the human voice of the passenger 16. The microphones 40 are arranged on a dashboard or a ceiling, and two microphones 40 can be provided, one for the passenger 16 in the driver's seat 12 and the passenger seat 13 and one for the passenger 16 in the rear seat 14. There are two methods of use for inputting the reproduced sound signal from the selector 34 in the conversation detection circuit 39. In the first usage method, the input of the playback sound signal from the selector 34 in the conversation detection circuit 39 is used to determine whether or not the source is being played back. That is, the conversation detection circuit 39 determines that the predetermined source is being reproduced during the period when the reproduction signal is input from the selector 34. The conversation detection circuit 39 sends a predetermined detection signal to the microcomputer 44 as a conversation period during reproduction during reproduction of a predetermined source and a voice input. In the second method of use, the input of the reproduced sound signal from the selector 34 in the conversation detection circuit 39 is obtained by subtracting the reproduced sound signal from the input of the sound detection signal from the microphone 40 to the conversation detection circuit 39 as a human voice. Used to detect. Since the sound signal detected by the microphone 40 is mixed with the reproduced sound in addition to the human voice, the microcomputer 44 can accurately detect the conversation period by receiving the difference between them from the conversation detection circuit 39. Note that the conversation detection circuit 39 may omit the input of the reproduction sound signal from the selector 34. The operation keys 46a, 46b,... 46n include a play (PLAY) key, a stop (STOP) key, a volume control (VOL) key, a sound field selection key, and the like, and are operated by the user. The microcomputer 44 controls the selector 34, the sound field control unit 37, and the volume / sound quality adjustment unit 49 based on inputs from the operation keys 46a, 46b,.

  The sound field control unit 37 generates a correction signal obtained by correcting the reproduction signal from the selector 34 so that a predetermined sound field is generated in the passenger compartment 11. The sound field generated by the sound field control unit 37 is a virtual sound image generated approximately at the ear level of the passenger 16 in the passenger compartment 11. In order to generate the virtual sound images of the left and right speakers 18a and 18b at positions different from the left and right speakers 18a and 18b, a well-known technique using a head related transfer function (for example, disclosed in Patent Document 3). Is used. When the microcomputer 44 issues an instruction to stop the sound field generation process to the sound field control unit 37, the sound field control unit 37 stops the correction process for the reproduction signal from the selector 34 and performs the reproduction. The signal is sent to the volume / sound quality adjustment unit 49 as it is. The volume / sound quality adjustment unit 49 controls the volume and quality of sound related to the audio signal from the sound field control unit 37 based on the control signal from the microcomputer 44, and sends it to the power amplification unit 50. Note that the selector 34, the sound field control unit 37, the volume / sound quality adjustment unit 49, and the power amplification unit 50 process audio signals related to the left and right sounds, that is, a total of two audio signals. When the reproduction signal from the medium is mono, that is, when there is only one reproduction signal, the sound field control unit 37 generates two reproduction signals using the same reproduction signal as the left and right reproduction signals. Then, process. The power amplifier 50 amplifies the left and right audio signals and sends them to the left and right speakers 18a and 18b.

  FIG. 4 is a functional block diagram of the in-vehicle acoustic device 55. The in-vehicle acoustic device 55 includes a plurality of speakers 56 arranged at different positions in the vehicle interior 11. The on-vehicle acoustic device 55 further includes a conversation period detection unit 58 and a reproduction signal correction unit 60. The conversation period detection means 58 detects whether it is a conversation period or a non-conversation period based on the utterance of the passenger 16. The reproduction signal correction means 60 is configured so that the virtual sound image generated by sound emission from the plurality of speakers 56 is almost at the ear level of the passenger 16 in the non-conversation period, and in the conversation period, A correction signal obtained by correcting the reproduction signal from the medium 63 is generated so that the virtual sound image disappears almost from the height of the ear of the passenger 16, and the correction signal is output to each speaker 56.

  The speaker 56 is, for example, the left and right speakers 18a and 18b in FIG. The medium 63 is a broadcast medium of CD 28, MD 29, DVD 30 or tuner 31. Typically, each medium 63 outputs reproduction signals for the left and right speakers 18a and 18b, that is, two reproduction signals. The reproduction signal correcting means 60 generates an output signal to the speaker 56 that sets the virtual sound image at an arbitrary position with respect to the actual sound image (= speaker 56) based on the head-related transfer function. The details of how to generate a virtual sound image at an arbitrary position are well known, for example, as described in Patent Document 3. According to the known method, a total of four head-related transfer functions Vt of complex numbers from the left and right virtual sound images to the left and right ears of one occupant 16 and the one occupant 16 from the actual sound image. A total of four head-related transfer functions Vr representing complex numbers up to the left and right ears are obtained in advance by experiments. In other words, in the experiment, a test speaker is placed at each of the virtual and real sound image positions, and a microphone attached to the ear of the dummy doll is placed at the ear position. From the relationship between the input audio signal of the test speaker and the detection signal of the microphone, A total of four head-related transfer functions (two left and right speakers × two left and right ears = 4) are obtained. Then, a total of four first correction signals obtained by multiplying the reproduction signal from the medium 63 by Vt / Vr are obtained, and the first correction signals for the left and right ears are added together to obtain one each on the left and right. That is, a total of two second correction signals are obtained, and the respective correction signals are input to the left and right speakers 56.

  The utterance of the passenger 16 can be detected by, for example, a microphone (not shown) whose directivity is adjusted toward the passenger 16. For example, the time when the utterance of the passenger 16 is detected for the first time is set as the start of the conversation period, and the time when a predetermined time elapses from when the utterance of the passenger 16 is detected at the end is set as the end of the conversation period. The fact that the virtual sound image disappears from the height of the ear of the passenger 16 includes that the virtual sound image is generated at a height that is not the height of the ear of the passenger 16. In this way, during the conversation period, the virtual sound image at almost the ear height of the passenger 16 disappears, so that the reproduced sound felt by the passenger 16 is weakened, and the passenger can smoothly advance the conversation.

  Preferably, the plurality of speakers 56 are provided at a height different from the height of the ears of the passenger 16. When the speaker 56 is the left and right speakers 18a and 18b, the left and right speakers 18a and 18b as actual sound images are arranged in the lower half of the left and right doors. The reproduction signal correction means 60 stops the reproduction signal correction process during the conversation period and outputs the reproduction signal as it is to the speaker 56 as a correction signal. As a result, during the conversation period, the sound image is the actual position of the left and right speakers 18a and 18b, so that the disturbance to the conversation due to the reproduced sound is suppressed.

  FIG. 5 is a flowchart of a method for controlling the in-vehicle acoustic device. The on-vehicle acoustic device includes a plurality of speakers 56 arranged at different positions in the passenger compartment 11. In S70, it is detected based on the voice of the passenger 16 whether it is a conversation period or a non-conversation period. If it is a non-conversation period, it will progress to S71, and if it is a conversation period, it will progress to S72. In S71, a correction signal obtained by correcting the reproduction signal from the medium 63 is generated so that the virtual sound image generated by sound emission from the plurality of speakers 56 is almost at the ear level of the passenger 16. The correction signal is output to each speaker 56. In S <b> 72, a correction signal obtained by correcting the reproduction signal from the medium 63 is generated so that the virtual sound image disappears almost from the height of the ear of the passenger 16, and the correction signal is output to each speaker 56. To do.

  During the non-conversation period, the virtual sound image is localized at almost the ear level of the occupant, so that the occupant 16 listens to the playback sound from the medium 63 at the volume set by the volume control unit. Become. On the other hand, during the conversation period, the virtual sound image at the height of the ear of the passenger disappears, the reproduced sound felt by the passenger 16 is weakened, and the passenger 16 can proceed with the conversation smoothly.

  Preferably, when the plurality of speakers 56 are the left and right speakers 18a and 18b, the plurality of speakers 56 are arranged at a height substantially different from the height of the ears of the passenger 16. . In S72, during the conversation period, the reproduction signal correction process is stopped, and the reproduction signal is output as it is to the speaker 56 as a correction signal.

  The disappearance of the virtual sound image of almost the ear height of the passenger in S72 includes that the virtual sound image is localized at a height different from the height of the passenger's ear. The plurality of speakers 56 may be the left and right speakers 18a and 18b arranged in the lower half of the left and right doors. In this case, in S72, the reproduction signal correction process is canceled and the reproduction signal is output to the speaker 56 as it is. As a result, the virtual sound image in the passenger compartment 11 disappears, and the passenger 16 hears the reproduced sound from the left and right speakers 18a and 18b as actual sound images, so the reproduced sound felt by the passenger 16 is weakened. The passenger 16 can proceed with the conversation smoothly.

  The control program of the present invention causes a computer to function as each unit of the in-vehicle acoustic device 55 of FIG. Alternatively, each step of FIG. 4 is executed by a computer.

It is a schematic side view of the motor vehicle which shows the example of arrangement | positioning of the speaker in a vehicle interior. It is a schematic plan view of the motor vehicle which shows the example of arrangement | positioning of the speaker in a vehicle interior. It is a hardware block diagram of a vehicle-mounted audio apparatus. It is a functional block diagram of a vehicle-mounted acoustic device. It is a flowchart of the control method of a vehicle-mounted audio apparatus.

Explanation of symbols

11: passenger compartment, 16: passenger, 55: vehicle-mounted acoustic device, 56: speaker, 58: conversation period detecting means, 60: reproduction signal correcting means, 63: media.

Claims (5)

In an in-vehicle audio apparatus provided with a plurality of speakers arranged at different positions in a vehicle interior and outputting reproduced sound of media,
Conversation period detection means for detecting a conversation period or a non-conversation period based on the utterance of the passenger, and a virtual sound image generated by the reproduced sound of the media from the plurality of speakers, Generates a correction signal that corrects the playback signal from the media so that the virtual sound image disappears from the height of the ear, so that the virtual sound image disappears from the height of the ear so that it becomes the height of the ear. Reproduction signal correction means for outputting the correction signal to each speaker,
A vehicle-mounted acoustic device characterized by comprising: The plurality of speakers are arranged at a height different from the height of the ears of the passenger,
2. The in-vehicle acoustic device according to claim 1, wherein the reproduction signal correcting means stops the reproduction signal correction processing during a conversation period and outputs the reproduction signal as it is to the speaker as a correction signal. . In a control method of an in-vehicle acoustic device provided with a plurality of speakers arranged at different positions in a vehicle interior and outputting reproduced sound of media,
A conversation period detection step for detecting whether the conversation period or non-conversation period based on the utterance of the passenger,
The virtual sound image generated by the reproduced sound of the media from the plurality of speakers is almost at the ear level of the passenger in the non-conversation period, and the virtual sound image of the passenger is in the conversation period. A reproduction signal correction step for generating a correction signal obtained by correcting the reproduction signal from the media and outputting the correction signal to each of the speakers so as to almost disappear from the height of the ear,
A control method for an on-vehicle acoustic device, comprising: The plurality of speakers are arranged at a height different from the height of the ears of the passenger,
4. The method of controlling an on-vehicle acoustic device according to claim 3, wherein in the reproduction signal correction step, the reproduction signal correction process is stopped during the conversation period, and the reproduction signal is directly output to the speaker as a correction signal. .   The program for functioning a computer as each means of the vehicle-mounted audio apparatus of Claim 1 or 2.

Images