JP2012231449A - Sound field generation device, sound field generation system, and sound field generation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a sound field generation device, a sound field generation system, and a sound generation method capable of appropriately generating a sound field in various environments.SOLUTION: A sound field generation device 2 according to the present invention comprises: a speaker part 30 which has a speaker of two or more channels and outputs acoustic signals toward a reflection part 40; a crosstalk cancellation operation part 20 which receives input of sound signals for which a sound image of the acoustic signals outputted by the speaker part 30 and reflected by the reflection part 40 to reach a listener 60 is localized as a virtual sound source at a position of the listener 60, and performs arithmetic processing, on the inputted sound signals, so as to cancel crosstalk between the respective channels of the acoustic signals reflected by the reflection part 40 to reach the listener 60 at the position of the listener 60; and a transfer function change part 29 which changes a transfer function used in the crosstalk cancellation operation part 20 according to the change of a sound field generation environment.

Description

  The present invention relates to a sound field generation device, a sound field generation system, and a sound field generation method, and in particular, a sound field generation device, a sound field generation system, and a sound field that generate a sound field using two or more channels of speakers. It relates to the generation method.

  Patent Document 1 discloses an in-vehicle speaker system for stereo reproduction for a listener who rides on a front part. In this in-vehicle speaker system, a pair of left and right speakers are provided on the instrument panel. The reproduced sound from the speaker is reflected on the windshield and reaches the listener in the driver's seat or passenger seat. The reproduced sound from the right speaker reaches the passenger seat listener, and the reproduced sound from the left speaker reaches the driver seat listener. Since the input signal is output from the speaker as it is, the sound field in front of the listener can be reproduced like a conventional stereo sound field, but the sound image localization in the rear of the listener cannot be reproduced.

  Patent Document 2 discloses a signal conversion device that converts a binaural signal that has been binaurally recorded. In Patent Document 2, a binaural signal is converted into a modified binaural signal so as to cancel out crosstalk that should occur in the reproduction sound field. The modified binaural signal is given to a plurality of speakers. Patent Document 3 discloses a playback system that plays back in a stereo dipole system. In the reproduction system of Patent Document 3, the spread angle of the loudspeaker pair is set in a range of 6 to 20 degrees.

JP-A-5-328483 JP 52-40101 A Special Table 2000-529106

  However, the method of Patent Document 3 has a problem that it cannot be applied to an environment such as the interior of an automobile. For example, when the stereo dipole method is used, even if a speaker is installed toward the ear, the influence of shielding by the handle and reflection by the side glass occurs. In particular, the interior of an automobile is an asymmetrical environment. For example, when the listener is a driver, the distance from the driver to the right side glass is shorter than the distance to the left side glass. Therefore, the influence of the reflection by the right side glass greatly appears on the driver's right ear, and the left and right reflections become non-uniform. Furthermore, it is not appropriate to place a speaker on the dashboard because placing the speaker on the dashboard to deliver sound directly to the listener will interfere with the driver's visual field. Furthermore, the sound field generation environment may change in an automobile or the like of an automated person.

  The present invention has been made in view of the above problems, and provides a sound field generation device, a sound field generation system, and a sound field generation method capable of appropriately generating a sound field even in various environments. With the goal.

A sound field generation device according to an aspect of the present invention includes a speaker that outputs sound of two or more channels, and is acoustically directed toward a reflection unit that reflects sound output from the speaker and transmits the sound to a listener. A speaker unit that outputs a signal, and an audio signal that is output from the speaker unit and reflected by the reflecting unit and arrives at the listener, the sound signal being localized as a virtual sound source at the position of the listener is input. From the speaker unit, the crosstalk between the channels of the acoustic signal reflected by the reflecting unit and reaching the listener with respect to the input audio signal is canceled at the position of the listener. A crosstalk canceling unit that performs arithmetic processing using a transfer function up to the listener, and a sound field generator including the reflecting unit and the listener as constituent elements In response to changes in the environment, the angle of the reflecting surface of the reflector portion, or the angle changing unit for changing at least one of the angle of the speaker and it is characterized in that it comprises. Thereby, since an audio | voice signal can be propagated in the direction suitable for a sound field production | generation environment, a sound field can be produced | generated appropriately.
A sound field generation device according to an aspect of the present invention includes a speaker that outputs sound of two or more channels, and is acoustically directed toward a reflection unit that reflects sound output from the speaker and transmits the sound to a listener. A speaker unit that outputs a signal, and an audio signal that is output from the speaker unit and reflected by the reflecting unit and arrives at the listener, the sound signal being localized as a virtual sound source at the position of the listener is input. From the speaker unit, the crosstalk between the channels of the acoustic signal reflected by the reflecting unit and reaching the listener with respect to the input audio signal is canceled at the position of the listener. A crosstalk canceling unit that performs arithmetic processing using a transfer function up to the listener, and a sound field generator including the reflecting unit and the listener as constituent elements In response to changes in the environment, it is characterized in that and a transfer function changing section for changing the transfer function used in the crosstalk cancellation unit. Accordingly, since the crosstalk can be canceled using a transfer function suitable for the sound field generation environment, the sound field can be generated appropriately.
In the above sound field generation device, the sound image is localized so that the sound image of the acoustic signal that has reached the listener becomes a virtual sound source at the position of the listener with respect to the audio signal input to the crosstalk cancellation unit. A sound image localization calculation unit for adding may further be provided, and a sound image of an audio signal input to the crosstalk cancellation unit may be localized by calculation by the sound image localization calculation unit.
In the above sound field generating device, the sound field generating device further includes a whispering voice conversion unit that converts an audio signal to be heard by the listener into a whispering voice, and the sound image localization calculation unit is output from the speaker unit and is reflected by the reflection unit and reflected by the listening unit. Calculation may be performed so that the sound image of the acoustic signal converted into the whispering voice that has reached the listener is localized as a virtual sound source in at least one ear of the listener.
In the above sound field generating device, the transfer function changing unit may change the transfer function according to a position of a seat on which the listener sits in the sound field generating environment.
In the above sound field generating device, the transfer function changing unit may change the transfer function according to an opening / closing position of a window provided in the sound field generating environment.
In the above sound field generating device, the transfer function changing unit changes the transfer function depending on whether a person other than the listener is sitting on a seat provided in the sound field generating environment. May be.
In the above sound field generation device, at least one of the angle of the reflection surface of the reflection unit or the angle of the speaker of the speaker unit is variable, and the seat on which the listener is seated in the sound field generation environment You may make it adjust at least one of the angle of the reflective surface of a front reflection part, or the angle of the said speaker according to a position.
In the above sound field generation device, the input sound signal may be a sound signal that is binaurally recorded at the position of the listener.
A sound field generation system according to an aspect of the present invention includes the sound field generation device described above and a reflection unit that reflects sound from the speaker unit toward the listener.
In the above sound field generation system, the reflection unit may be a vehicle body or an in-vehicle device.
In the above sound field generation system, the reflection unit is at least one of a windshield, a side glass, and a roof glass of an automobile, and the transmission coefficient changing unit is at least one of the windshield, the side glass, and the roof glass. The transfer function may be changed according to one open / close position.
In the above sound field generation system, when the side glass closer to the listener is opened from a predetermined position among the left and right side glasses of the automobile, the side glass closer to the listener is used as a reflection part. The other glass is used as a reflection part to deliver the acoustic signal to the listener, and when the side glass closer to the listener is closed than a predetermined position, the side glass closer to the listener is reflected to the reflection part. The angle of the speaker may be changed so that the acoustic signal is delivered to the listener.
In the sound field generation method according to one aspect of the present invention, a sound image of an acoustic signal output from a speaker unit having a speaker that outputs sound of two or more channels and reflected by a reflecting unit and reaches the listener is a position of the listener. And inputting a sound signal localized as a virtual sound source in the sound source, and crosstalk between the channels of the sound signal reflected by the reflection unit and reaching the listener for the input sound signal. A step of performing arithmetic processing using a transfer function from the speaker unit to the listener so as to be canceled at the position of the listener, and transmitting the sound to the listener by reflecting the sound output from the speaker unit A step of outputting an acoustic signal toward the reflection unit, and according to a change in a sound field generation environment including the reflection unit and the listener as components , In which and a step of changing the transfer function used in the arithmetic processing of canceling the crosstalk.

  As described above, according to the present invention, it is possible to provide a sound field generation device, a sound field generation system, and a sound field generation method that can appropriately generate a sound field even in various environments.

It is a block diagram which shows the structure of the sound field generation system concerning Embodiment 1 of this invention. It is a figure which shows the whole structure of the sound field production | generation system concerning Embodiment 1 of this invention. It is a block diagram which shows a sound image localization calculating part. It is a figure which shows a crosstalk cancellation calculating part. 5 is a top view showing an installation example 1 of Embodiment 1. FIG. 6 is a side view showing an installation example 1 of Embodiment 1. FIG. It is a side view which shows the state which moved the driver's seat back. 6 is a top view showing an installation example 2 of Embodiment 1. FIG. FIG. 6 is a top view showing an installation example 3 of the first embodiment. 6 is a top view showing an installation example 4 according to Embodiment 1. FIG. FIG. 6 is a top view showing an installation example 5 of the first embodiment. FIG. 10 is a top view showing an installation example 6 of the first embodiment. FIG. 10 is a perspective view showing a left speaker and a left reflecting portion used in Installation Example 7 of Embodiment 1. FIG. 10 is a top view showing an installation example 7 of the first embodiment. 10 is a perspective view showing a speaker unit and a reflection unit used in Installation Example 8 of Embodiment 1. FIG. FIG. 10 is a top view showing an installation example 8 according to the first embodiment. FIG. 10 is a top view showing an installation example 9 according to the first embodiment. FIG. 10 is a side view showing an installation example 9 according to the first embodiment. It is a block diagram which shows the structure of the sound field generation system concerning Embodiment 2 of this invention. It is a figure which shows the whole structure of the sound field production | generation system concerning Embodiment 2 of this invention. It is a figure which shows typically the recording site which performs binaural recording using a dummy head. It is a block diagram which shows the structure of the speaker unit part of the sound field generation system concerning Embodiment 3 of this invention. It is a figure which shows typically the state currently seated in the front passenger seat and rear seat of a motor vehicle. It is a block diagram which shows the structure of the speaker unit part of the sound field generation system concerning Embodiment 4 of this invention. It is a block diagram which shows the structure of the speaker unit part of the sound field generation system concerning Embodiment 5 of this invention. It is a top view which shows a mode that the sound field production | generation system concerning Embodiment 5 of this invention was installed in the motor vehicle. It is a top view which shows the state which opened the right side glass. It is a side view which shows the opening / closing position of the side glass which switches the direction of a right speaker. It is a top view which shows a mode that the sound field production | generation system concerning Embodiment 6 of this invention was installed in the motor vehicle. It is a top view which shows the state which opened the roof glass. It is a top view which shows the asymmetrical arrangement | positioning of a speaker. It is a side view which shows the asymmetrical arrangement | positioning of a speaker. It is a block diagram which shows the structure of the sound field generation system concerning Embodiment 7 of this invention. It is a figure which shows the whole structure of the sound field generation system concerning Embodiment 7 of this invention.

Embodiment 1
Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing the configuration of the sound field generation system according to the first embodiment. FIG. 2 is a diagram schematically showing the overall configuration of the sound field generation system. The sound field generation system 1 includes a sound image localization calculation unit 10, a crosstalk cancellation calculation unit 20, a speaker unit 30, a reflection unit 40, a sheet 110, and a sheet sensor 111. The sound image localization calculation unit 10, the crosstalk cancellation calculation unit 20, and the speaker unit 30 constitute the sound field generation device 2. In the present embodiment, the speaker unit 30 in the sound field generation system 1 has two speakers arranged close to each other. That is, the speaker unit 30 includes a left speaker 31 and a right speaker 32.

  An input signal p (t) that is an audio signal is input to the sound image localization calculation unit 10. In the present embodiment, description will be made assuming that the input signal p (t) is a monaural signal. The input signal p (t) is input to the sound field generating device 2 by reproducing a CD (Compact Disc), a DVD (Digital Versatile Disc), an MP3 player, or the like. The sound image localization calculation unit 10 performs a left ear stereo signal Ls obtained by performing calculation processing for appropriately locating the virtual sound source p ′ (t) at the listening position of the listener 60 with respect to the input signal p (t). And a stereo signal Rs for the right ear can be obtained. That is, two-channel stereo signals Ls and Rs are generated. In the present embodiment, the sound field generation system that outputs two-channel audio will be described. However, the number of channels may be two or more.

  The crosstalk cancellation calculation unit 20 performs a crosstalk cancellation process on the stereo signals Ls and Rs based on the input audio signal. As a result, speaker signals Lsp and Rsp can be obtained as audio signals whose crosstalk has been canceled at the ears of the listener 60. That is, the influence on the right ear 62 due to the sound emitted from the left speaker 31 and the influence on the left ear 61 due to the sound emitted from the right speaker 32 can be reduced. The crosstalk when the sound signal is actually output from the speaker unit 30 is canceled, and the sound signal that reaches the left ear 61 via the sound field is processed so as to be the same sound signal as the stereo signal Ls. The sound signal that reaches the right ear 62 via is processed so that it becomes the same sound signal as the stereo signal Rs.

  The sound image localization calculation unit 10 and the crosstalk cancellation calculation unit 20 are configured by, for example, FIR (Finite Impulse Response) filter processing by a DSP (Digital Signal Processor).

  The speaker unit 30 is a stereo speaker and includes a left speaker 31 and a right speaker 32. The speaker unit 30 emits an acoustic signal based on the speaker signals Lsp and Rsp in which the sound image is localized with respect to the listener 60 and the crosstalk is canceled. That is, the left speaker 31 outputs an acoustic signal based on the speaker signal Lsp, and the right speaker 32 outputs an acoustic signal based on the speaker signal Rsp. The speaker signal Lsp includes a cancel component for canceling the influence of the sound from the right speaker 32 on the left ear 61. The speaker signal Rsp includes the sound from the left speaker 31 to the right ear 62. A cancel component for canceling the influence is included.

  The reflection unit 40 is disposed to face the sound emission surface of the speaker unit 30. The left speaker 31 and the right speaker 32 output acoustic signals corresponding to the speaker signals Lsp and Rsp toward the reflection unit 40. The reflection unit 40 includes a left reflection unit 41 and a right reflection unit 42. The left reflection unit 41 reflects the acoustic signal radiated from the left speaker 31 toward the left ear 61 of the listener 60. The right reflection unit 42 reflects the acoustic signal output from the right speaker 32 toward the right ear 62 of the listener 60. As the left reflecting portion 41 and the right reflecting portion 42, a wall surface or a window glass in an acoustic signal output environment can be used. Further, the left reflection unit 41 and the right reflection unit 42 may be provided in the speaker unit 30.

  The seat 110 is a seat on which the listener 60 is seated, and the seat sensor 111 detects the seat position of the seat 110. From the seat position detected by the seat sensor 111, the position of the ear of the listener 60 who has boarded the seat sensor 111 is estimated. The sheet sensor 111 outputs a detection signal indicating the sheet position to the crosstalk cancellation calculation unit 20, and the crosstalk cancellation calculation unit 20 changes the transfer function according to the sheet position, as will be described later.

  As described above, the crosstalk is canceled so that the component of the stereo signal Ls in the acoustic signal output from the left speaker 31 does not affect the right ear 62. Similarly, the crosstalk is canceled so that the stereo signal Rs component of the acoustic signal output from the right speaker 32 does not affect the left ear 61. Therefore, the signals Le and Re reaching the left ear 61 and the right ear 62 of the listener 60 are in a state in which crosstalk is canceled. Therefore, a sound field similar to the sound field at the position of the virtual listener 80 can be generated.

  In the present embodiment, the sound field is generated as described above. Since the sound image localization calculation unit 10 performs calculation so that the sound image is localized at the position of the listener, the listener 60 can localize an appropriate virtual sound source. Furthermore, since the crosstalk is canceled, the listener 60 can listen as if the stereo signal Rs and the stereo signal Ls are output from the virtual sound source. Thereby, the listener 60 can listen to a three-dimensional sound field. In addition, the sound from the speaker unit 30 reaches the listener 60 via the reflection unit 40. By disposing the left speaker 31 and the right speaker 32 toward the reflecting unit 40, direct sound from the left speaker 31 and the right speaker 32 is prevented from reaching the listener 60. For example, the acoustic signal output from the left speaker 31 can be prevented from reaching the left ear 61 directly. Disturbance of the sound field due to a mixture of a direct sound that reaches the listener 60 without passing through the reflector 40 and a reflected sound that reaches the listener 60 via the reflector 40 can be reduced.

  Further, since the crosstalk cancellation calculation unit 20 changes the transfer function according to the position of the seat 110 on which the listener 60 sits, the crosstalk is generated with the transfer function according to the relative position between the speaker unit 30 and the listener 60. Canceled. Thereby, a transfer function can be optimized and a sound field can be generated more appropriately.

  Next, processing in the sound image localization calculation unit 10 will be described. FIG. 3 is a diagram schematically showing processing in the sound image localization calculation unit 10. The sound image can be localized by performing a convolution operation on the input signal p (t), which is a monaural signal input to the sound image localization calculation unit 10, by the sound image localization calculation unit 10. Specifically, the sound image localization calculation unit 10 includes convolution calculation units 11 and 12. The convolution operation unit 11 can obtain the stereo signal Ls by performing the convolution operation of the input signal p (t) and the transfer function bl (t). Similarly, the convolution operation unit 12 can obtain the stereo signal Rs by performing the convolution operation of the input signal p (t) and the transfer function br (t). Specifically, Ls (t) = p (t) * bl (t) and Rs (t) = p (t) * br (t). bl (t) and br (t) are set according to the position of the sound image to be reproduced. The stereo signals Ls and Rs calculated by the sound image localization calculation unit 10 are given a time difference, a volume difference and the like according to the position of the sound image.

  Next, processing in the crosstalk cancellation calculation unit 20 will be described with reference to FIG. The crosstalk cancellation processing in the crosstalk cancellation calculation unit 20 uses, for example, the crosstalk cancellation processing described in Japanese Patent Laid-Open No. 52-40101. As shown in FIG. 2, the transfer function (transfer characteristic) from the left speaker 31 to the left ear 61 is a11, the transfer function from the left speaker 31 to the right ear 62 is a21, and the right speaker 32 to the left ear 61 is The transfer function is a12, and the transfer function from the right speaker 32 to the right ear 62 is a22. Thus, the transfer functions a11, a12, a21, and a22 are filter coefficients indicating the transmission characteristics from the left and right speakers to the left and right ears. The transfer function a11 is set according to the position of the left ear 61 with respect to the left speaker 31, and the transfer function a21 is set according to the position of the right ear 62 with respect to the left speaker 31. Similarly, the transfer function a12 is set according to the position of the left ear 61 with respect to the right speaker 32, and the transfer function a22 is set according to the position of the right ear 62 with respect to the right speaker 32. The transfer functions a11, a12, a21, and a22 are, for example, functions of frequency.

  The crosstalk cancellation calculation unit 20 sets transfer functions a11, a12, a21, and a22 that cancel the wraparound from the left speaker 31 to the right ear 62 and the wraparound from the right speaker 32 to the left ear 61. As the transfer functions a11, a12, a21, and a22, the influence reflected by the reflecting unit 40 is taken into consideration. That is, the transfer function is set so that the influence of the sound reflected by the reflecting unit 40 around the other ear is canceled.

  The crosstalk cancellation calculation unit 20 includes filters 21 and 22, delay units 23 and 24, inverters 25 and 26, and adders 27 and 28. Here, the filters 21 and 22 and the delay units 23 and 24 have the following filter coefficients.

Filter 21: a11 / (a11 · a22-a12 · a21)
Filter 22: a22 / (a11 · a22-a12 · a21)
Delay device 23: a12 / a11
Delay device 24: a21 / a22

Accordingly, the speaker signals Lsp and Rsp are as follows.
Lsp = (Ls · a11−Rs · a21) / (a11 · a22−a12 · a21)
Rsp = (Rs · a22−Ls · a12) / (a11 · a22−a12 · a21)

  The crosstalk cancellation calculation unit 20 calculates speaker signals Lsp and Rsp in which the crosstalk is canceled from the stereo signals Ls and Rs using a transfer function. As described above, the crosstalk cancellation calculation unit 20 performs a process of canceling the crosstalk of the sound that occurs when the speaker signals Ls and Rs in which the sound image is localized are output from the speaker unit 30.

  Furthermore, a transfer function changing unit 29 is provided in the crosstalk cancellation calculating unit 20. The transfer function changing unit 29 receives a detection signal from the sheet sensor 111 and switches the transfer function according to the sheet position. Specifically, the transfer function changing unit 29 includes storage means such as a memory, and stores a plurality of transfer functions a11, a12, a21, and a22. Assuming that the four transfer functions a11, a12, a21, and a22 are one set, the transfer function changing unit 29 stores a plurality of sets of transfer functions and changes the transfer functions according to the sheet position. When the sheet sensor 111 detects that the sheet position has been changed, the transfer function changing unit 29 switches the transfer function, so that the crosstalk cancellation calculation unit 20 changes the transferred transfer functions a11, a12, a21, and a22. To perform crosstalk cancellation processing.

  Hereinafter, an installation example in which the sound field generation system 1 is installed will be described. In the following installation example, the interior of the automobile 100 is the installation environment of the sound field generation system 1. In addition, the reflection part 40 in the sound field generation system 1 is a vehicle body in the vehicle 100 or an in-vehicle device attached to the vehicle body, and is an object having a reflective surface. If it is a vehicle body, it may be a windshield, a side glass, a roof glass, a dashboard, a room mirror, or the like, and if it is an in-vehicle device, it may be a car navigation, a drive recorder, or the like. In addition, contents common to the following description of the plurality of installation examples will be omitted as appropriate.

Installation example 1.
The installation example 1 of said sound field generation system 1 is demonstrated based on FIGS. FIG. 5 is a diagram illustrating a state in which the sound field generation system 1 is installed, and is a diagram of the main part of the automobile 100 as viewed from above. In FIG. 5, the upper side of the page is the front of the automobile 100. 6 and 7 are diagrams showing the arrangement of the speaker unit 30 and the reflecting unit 40, and are views showing the installation state from the side. 6 and 7, the left side of the page is the front of the automobile 100. The automobile 100 is provided with a windshield 101, a right side glass 102, a left side glass 103, and a driver's seat 106. 5 to 7, a part of the configuration of the automobile 100 is omitted.

  In this installation example 1, the left speaker 31 and the right speaker 32 are installed in a state where they are embedded in the dashboard 109, and the left speaker 31 and the right speaker 32 are installed side by side. Further, the windshield 101 of the automobile 100 serves as the reflecting section 40, and the left speaker 31 and the right speaker 32 radiate acoustic signals toward the windshield 101 serving as the reflecting section 40. Then, the acoustic signal reflected by the reflecting unit 40 reaches the listener 60 who has boarded the driver's seat 106. By adopting such an installation form, it is possible to prevent the influence of the shielding object such as the handle 105 due to the direct sound of the acoustic signal output from the speaker unit 30 and the influence of the right side glass 102 and the left side glass 103 due to other reflected sounds. Therefore, the driver's view is not obstructed by the installation of the speaker.

  Further, as shown in FIG. 6, it is preferable that the speaker unit 30 is arranged with the output direction of the acoustic signal from the speaker unit 30 facing upward. For example, the output direction of the acoustic signal from the speaker unit 30 is set vertically upward, and the speaker unit 30 is disposed immediately below a location where the windshield 101 is inclined 45 degrees from the vertical direction. By doing so, the acoustic signal reflected by the reflecting unit 40 propagates in the horizontal direction and reaches the listener 60. Furthermore, by making the position where the acoustic signal enters the reflecting portion 40 higher than the handle 105, it is possible to prevent sound shielding and reflection by the handle 105. Therefore, an appropriate sound image can be generated for the listener 60. As a form of the speaker unit 30, it can be installed as a part of the head-up display unit. In addition, the inclination angle of the reflection surface of the reflection unit 40 and the output direction of the acoustic signal from the speaker unit 30 are not particularly limited. The reflection surface of the reflection unit 40 may be a flat surface or a curved surface. Note that the sound image localization calculation unit 10 and the crosstalk cancellation calculation unit 20 preferably perform calculations corresponding to the shape of the reflection surface of the reflection unit.

  By installing in this way, the acoustic signal that is reflected by the reflecting portion 40 and reaches the listener 60 reaches the listener 60 without being affected by a shield such as the handle 105. Furthermore, since the sound reflected by the reflection unit 40 from the speaker unit 30 reaches the listener 60, it is possible to prevent the sound field from being disturbed by mixing other unnecessary reflection sounds. Therefore, an appropriate sound field can be generated at the position of the listener 60. The listener 60 is not limited to the driver, and may be a passenger who has boarded the passenger seat.

  In this installation example, the driver's seat 106 is the seat 110 shown in FIGS. 1 and 2, and the transfer function changing unit 29 detects the change of the seat position of the driver's seat 106 by the seat sensor 111, so that the transfer function changing unit 29 To change. For example, if the driver's seat 106 is slid in the direction of the arrow from the seat position shown in FIG. 6, the seat position of the driver's seat 106 shifts backward as shown in FIG. The positional relationship between the part 30 and the reflecting part 40 changes. Accordingly, since the distance of the acoustic signal emitted from the speaker unit 30 reaching the left ear 61 and the right ear 62 of the listener 60 changes, the transfer functions a11, a12, a21, and a22 are switched to different sets. Thus, a sound field can be generated more appropriately by changing the transfer function according to the sound field generation environment. Note that the transfer function may be switched stepwise according to the sheet position or continuously. When the transfer function is switched in stages, the transfer function is the same up to a certain range where the seat position of the driver's seat 106 is, and when the range is exceeded, the transfer function is changed. When the transfer function is continuously switched, for example, the transfer function is a function having the coordinates of the sheet position as a variable.

  Thus, when the position of the driver's seat 106 on which the listener 60 sits changes, the positional relationship between the left speaker 31, the right speaker 32, the left ear 61, and the right ear 62 changes. If an appropriate transfer function is prepared in advance according to the seat position and the transfer function changing unit 29 switches and uses the transfer function according to the seat position, a sound field can be generated more appropriately. In the above description, the listener 60 is the driver who sits in the driver's seat 106. However, a passenger sitting in the passenger seat or the rear seat may be the listener 60. In this case, too, the passenger seat or the rear seat is seated. The transfer function changing unit 29 changes the transfer function according to the position.

  The method for detecting the sheet position is not particularly limited. For example, when the driver's seat 106 is an electric seat, the seat position can be detected by detecting the rotational speed and rotational speed of a motor or the like that moves the seat. When the driver's seat 106 is a manually moving seat, a position sensor that serves as the seat sensor 111 is provided in a guide mechanism or the like that guides the sliding movement of the seat. Of course, the transfer function changing unit 29 may switch the transfer function in accordance with not only the sliding movement of the seat in the front-rear direction but also the inclination position of the seat back portion.

  Furthermore, it is also possible to measure a transfer function in consideration of the influence of the reflection unit 40 using a dummy head. In an environment where the sound field generation system 1 is actually installed or in substantially the same environment, a dummy head is arranged at the head position when the listener 60 is seated. The impulse response is measured for each of the left speaker 31 and the right speaker 32. That is, when an impulse signal is output from the left speaker 31 in a state where dummy heads with microphones provided on the left and right ears are arranged in the driver's seat 106, the impulse signals reflected by the left reflector 41 are converted to the left and right microphones of the dummy head. To reach. Similarly, the impulse response of the right speaker 32 is measured with the left and right microphones. Thus, a transfer function corresponding to the sound field generation environment can be obtained from the signal acquired by the microphone. By performing the same impulse response measurement by changing the seat position, it is possible to set an appropriate transfer function according to the seat position, and even when there is a change in the sound field generation environment, the sound is more appropriately A field can be created.

  By doing in this way, the transfer function which considered the reflection by the reflection part 40 can be calculated | required, and crosstalk can be canceled reliably by using this transfer function. For this reason, a more appropriate sound field can be generated. In the interior of the automobile 100, a sound absorbing surface that absorbs sound and a reflecting surface that reflects sound are mixed, but the acoustic signal is delivered to the ear of the listener 60 via the reflecting portion 40 as in this installation example. Thus, the influence of other reflections in the sound field generation space can be reduced. Therefore, an appropriate sound field can be generated even in various environments.

Installation example 2.
Next, installation example 2 will be described with reference to FIG. FIG. 8 is a view showing a state in which the sound field generation system 1 is installed, and is a view of the automobile 100 as viewed from above. In the installation example 2, the description of the same configuration as the installation example 1 is omitted as appropriate. In the installation example 2, the left speaker 31 and the right speaker 32 are installed side by side on the dashboard 109. The right side glass 102 and the left side glass 103 are used as a right reflecting portion 42 and a left reflecting portion 41, respectively. For example, the right speaker 32 outputs an acoustic signal toward the right side glass 102, so that the right side glass 102, which is the right reflection unit 42, receives the acoustic signal from the right speaker 32 on the driver's seat 106. Reflected toward the right ear 62 of the camera. The left speaker 31 outputs an acoustic signal toward the left side glass 103, so that the left side glass 103, which is the left reflection unit 41, reflects the acoustic signal from the left speaker 31 toward the left ear 61 of the listener 60. To do. Further, the transfer function changing unit 29 changes the transfer function according to the seat position of the driver's seat 106. By setting it as such an installation example, the effect similar to each said installation example can be acquired.

Installation example 3.
Next, installation example 3 will be described with reference to FIG. FIG. 9 is a view showing a state where the sound field generation system 1 is installed, and is a view of the automobile 100 as viewed from above. In the installation example 3, the description of the same configuration as in the installation example is omitted as appropriate. A roof glass 104 serving as a sunroof is provided on the ceiling of the automobile 100. The roof glass 104 is provided immediately above the driver's seat 106. In the installation example 3, the roof glass 104 is used as the reflecting portion 40. For example, the left speaker 31 and the right speaker 32 are installed side by side on the dashboard 109, and the left speaker 31 and the right speaker 32 output an acoustic signal toward the roof glass 104, thereby reflecting the reflection unit. The roof glass 104 that is 40 reflects the acoustic signals from the left speaker 31 and the right speaker 32 toward the left ear 61 and the right ear 62 of the listener 60 who has boarded the driver's seat 106. Further, the transfer function changing unit 29 changes the transfer function according to the seat position of the driver's seat 106. By setting it as such an installation example, the effect similar to each said installation example can be acquired.

Installation example 4.
Next, installation example 4 will be described with reference to FIG. FIG. 10 is a diagram illustrating a state in which the sound field generation system 1 is installed, and is a diagram of the automobile 100 as viewed from above. In the installation example 4, the description of the same configuration as the installation example is omitted as appropriate. A roof glass 104 serving as a sunroof is provided on the ceiling of the automobile 100. In the installation examples 1 to 3, the driver who has boarded the driver's seat 106 is the listener 60, but in this installation example 4, the passenger who has boarded the rear seat 107 is the listener 60. In the installation example 4, the roof glass 104 is used as the reflecting portion 40. For example, the left speaker 31 and the right speaker 32 are installed side by side on the dashboard 109, and the left speaker 31 and the right speaker 32 output acoustic signals toward the roof glass 104. The roof glass 104 serving as the reflection unit 40 reflects the acoustic signals from the left speaker 31 and the right speaker 32 toward the left ear 61 and the right ear 62 of the listener 60. Further, the transfer function changing unit 29 changes the transfer function according to the seat position of the rear seat 107. By setting it as such an installation example, the effect similar to each said installation example can be acquired.

Installation example 5.
Next, installation example 5 will be described with reference to FIG. FIG. 11 is a view showing a state in which the sound field generation system 1 is installed, and is a view of the automobile 100 as viewed from above. In the installation example 5, the description of the same configuration as the installation example is omitted as appropriate. In the present installation example 5, as in the installation examples 1 to 3, the driver who has boarded the driver's seat 106 is the listener 60. In the installation example 5, the left speaker 31 and the right speaker 32 are installed side by side on the dashboard 109, and the front glass 101 and the right side glass 102 are used as the reflecting portion 40. The left speaker 31 outputs an acoustic signal toward the windshield 101, so that the windshield 101 as the reflection unit 40 reflects the acoustic signal from the left speaker 31 toward the left ear 61 of the listener 60. . The right speaker 32 outputs an acoustic signal toward the right side glass 102, so that the right side glass 102, which is the reflection unit 40, transmits the acoustic signal from the right speaker 32 to the right ear of the listener 60 who has boarded the driver's seat 106. Reflects toward 62. Further, the transfer function changing unit 29 changes the transfer function according to the seat position of the driver's seat 106. By setting it as such an installation example, the effect similar to each said installation example can be acquired.

  In this installation example, the front glass 101 is used as the reflecting portion 40 for one ear (here, the left ear 61), and the side glass is used as the reflecting portion 40 for the other ear (here, the right ear 62). In particular, it is preferable that the side glass is used as the reflecting portion 40 in the ear closer to the side glass (right ear 62), and the windshield 101 is used as the reflecting portion in the ear far from the side glass (left ear 61). Thereby, the influence of a direct sound can be reduced. Of course, in the case of a left-hand drive vehicle, the left and right are reversed. That is, in the case of a left-hand drive vehicle, the left side glass 103 is the left reflecting portion 41 and the windshield 101 is the right reflecting portion 42. Alternatively, when the passenger on the passenger seat is the listener 60, the left side glass 103 is the left reflecting portion 41 and the windshield 101 is the right reflecting portion 42. Of course, with respect to one ear, as shown in the installation example 3, instead of the windshield 101, a roof glass may be used as the reflecting portion 40.

Installation example 6.
Next, installation example 6 will be described with reference to FIG. FIG. 12 is a view showing a state in which the sound field generation system 1 is installed, and is a view of the automobile 100 as viewed from above. In the installation example 6, the description of the same configuration as the installation example 1 is omitted as appropriate. In the installation example 6, the left speaker 31 and the right speaker 32 are installed side by side on the dashboard 109. As a specific example, the left speaker 31 is disposed on the front side, that is, the back side of the dashboard 109, the right speaker 32 is disposed on the rear side, that is, the front side of the dashboard 109, and the right side glass 102 is used as the reflecting portion 40. Yes. The right speaker 32 outputs an acoustic signal toward the right side glass 102, whereby the right side glass 102, which is the right reflection unit 42, transmits the acoustic signal from the right speaker 32 to the right of the listener 60 who has boarded the driver's seat 106. Reflected toward the ear 62. The left speaker 31 outputs an acoustic signal toward the right side glass 102, so that the right side glass 102, which is the left reflection unit 41, reflects the acoustic signal from the left speaker 31 toward the left ear 61 of the listener 60. To do. Further, the transfer function changing unit 29 changes the transfer function according to the seat position of the driver's seat 106. By setting it as such an installation example, the effect similar to each said installation example can be acquired.

Installation example 7.
Next, installation example 7 will be described with reference to FIGS. 13 and 14. FIG. 13 is a perspective view showing the left speaker 31 and the left reflecting portion 41. FIG. 14 is a view showing a state where the sound field generation system 1 is installed, and is a view of the automobile 100 as viewed from above. In the present installation example 7, various glasses of the automobile 100 are not used as the reflection unit 40, and the reflection unit 40 is separately installed on the dashboard 109. As shown in FIG. 13, the angle of the reflection surface of the left reflection unit 41 is variable, and the reflection unit 40 adjusts the mounting angle of the left reflection unit 41 so that the acoustic signal output from the left speaker 31 is adjusted. The reflection direction can be changed. Similarly, the angle of the reflecting surface of the right reflecting portion 42 is variable. When the angle of the reflecting surface of the reflecting unit 40 is changed, the transfer function changing unit 29 switches the transfer function.

  In the example 7 of installation, the speaker part 30 and the reflection part 40 are installed in the dashboard 109 as shown in FIG. The left reflection unit 41 is disposed in front of the left speaker 31, the right reflection unit 42 is disposed in front of the right speaker 32, and the reflection angles of the left reflection unit 41 and the right reflection unit 42 are variable. Yes. In the example shown in FIG. 14, the listener 60 is a passenger on the passenger seat 108. In addition, although the reflection angle of the left reflection part 41 and the right reflection part 42 is good also as variable integrally, in order to produce | generate a more suitable sound field, it is preferable to make it variable independently. Further, the reflection angle may be variable in all of the vertical direction, the horizontal direction, and the front-rear direction, or may be variable only in an arbitrary direction.

  In this installation example, the listener 60 can be switched by adjusting the reflection angle of the reflection unit 40. In other words, in FIG. 14, the listener 60 is a passenger who has boarded the passenger seat 108, but by changing the reflection angle of the speaker unit 30 from the state shown in FIG. Can reach the person. The reflection angle can be adjusted automatically using a drive mechanism such as a motor. In this case, a reflection angle suitable for each listener 60 may be set in advance, and a motor or the like may be driven to change the listener 60 to a preset reflection angle. Of course, the passenger may manually change the reflection angle of the reflector 40. By doing in this way, since a reflection angle can be adjusted according to a passenger, more suitable sound field production | generation is attained with respect to each listener 60. FIG. Further, the reflection surface of the reflection unit 40 may be a flat surface or a curved surface.

  By setting it as such an example, since the influence of obstructions, such as the handle | steering-wheel 105, can be reduced and it can prevent that the direct sound from the speaker part 30 reaches the listener 60, it is in the reflection part 40. By mixing the direct sound that reaches the listener 60 without being reflected and the reflected sound that is reflected by the reflector 40 and reaches the listener 60, it is possible to prevent the sound field from being disturbed. Therefore, a sound field can be generated appropriately. Further, when the listener 60 is switched, the transfer function changing unit 29 changes the transfer function, so that an appropriate sound field can be generated for each listener 60. Further, when the listener 60 is the passenger seat 108, the transfer function is changed according to the seat position of the passenger seat 108, and when the listener 60 is the driver seat 106, the transfer function is changed according to the seat position of the driver seat 106. By doing so, a sound field can be generated more appropriately.

  In the above description, the angle of the reflection surface of the reflection unit 40 is changed. However, the same effect can be obtained by changing the angle of the acoustic radiation surface of the speaker unit 30. Furthermore, when changing the angle of the acoustic radiation surface of the speaker unit 30, the right side glass 102, the left side glass 103, the front glass 101, and the roof glass 104 can be used as the reflection unit 40 as shown in the installation examples 1 to 7. .

Installation example 8.
Next, installation example 8 will be described with reference to FIG. FIG. 15 is a diagram illustrating the left speaker 31, the right speaker 32, the left reflection unit 41, and the right reflection unit 42 in which the sound field generation system 1 is installed. In this installation example, the left speaker 31, the right speaker 32, the left reflecting portion 41, and the right reflecting portion 42 are embedded in the dashboard 109. That is, the dashboard 109 is provided with two embeddings, and the left speaker 31 and the left reflecting portion 41 are integrally disposed in one embedding, and the right speaker 32 and the right reflecting portion 42 are disposed in the other embedding. Arranged integrally. Therefore, the wall of the embedded portion provided in the dashboard 109 functions as the left reflecting portion 41 and the right reflecting portion 42. In other words, in this installation example, the windshield 101, the right side glass 102, the left side glass 103, and the roof glass 104 are not used as the reflecting portion 40. By doing in this way, the sound field production | generation apparatus 2 can be installed easily. Of course, the left speaker 31, the right speaker 32, the left reflection unit 41, and the right reflection unit 42 may all be disposed in one embedding.

  FIG. 16 shows a state in which the configuration shown in FIG. FIG. 16 is a diagram illustrating a state in which the sound field generation system 1 is installed, and is a diagram of the automobile 100 as viewed from above. As shown in FIG. 16, by adjusting the angles of the left speaker 31 and the right speaker 32, the output direction of the acoustic signal from the left speaker 31 and the right speaker 32 can be changed, and the acoustic signal is transmitted through the reflector 40. It is possible to change the arrival position reached via Note that the angles of the acoustic radiation surfaces of the left speaker 31 and the right speaker 32 may be adjusted independently.

  By changing the direction of the left speaker 31 and the right speaker 32 and changing the output direction of the left speaker 31 and the right speaker 32 according to the seat position, an acoustic signal is output in an optimum direction even when the seat position changes. can do. Further, when the seat position changes, the transfer function changing unit 29 switches the transfer function, so that a sound field can be generated more appropriately. Of course, not only the left speaker 31 and the right speaker 32 but also the directions of the left reflecting portion 41 and the right reflecting portion 42 may be further variable. That is, if the direction of at least one of the speaker unit 30 and the reflection unit 40 is made variable, the output direction of the acoustic signal can be changed according to the seat position. Further, the listener 60 may be switched by changing the direction of the speaker unit 30 as in the case of the installation example 7.

Installation example 9.
Next, installation example 9 will be described with reference to FIG. FIG. 17 is a diagram showing the left speaker 31, the right speaker 32, the car navigation body 117, and the car navigation display surface 118 in which the sound field generation system 1 is installed. In this installation example, the left speaker 31, the right speaker 32, and the car navigation body 117 are embedded in the dashboard 109. That is, the dashboard 109 is provided with three embeddings, one is used for embedding the car navigation body 117, and the other two are present behind the car navigation display surface 118, and the left speaker 31. And used for embedding the right speaker 32. Then, the acoustic signal of the left speaker 31 is reflected toward the left ear 61 of the listener 60 with the car navigation display surface 118 as a reflecting surface. Similarly, the sound signal of the right speaker 32 is reflected toward the right ear 62 of the listener 60 with the car navigation display surface 118 as a reflection surface. Further, since the transfer function changing unit 29 is switched to the transfer function using the car navigation display surface 118, the sound field can be generated more appropriately.

  Further, as shown in FIG. 18, the speaker unit 30 is embedded in a space located behind the car navigation display surface 118. The speaker unit 30 is installed with a speaker angle such that an acoustic signal output from the speaker unit 30 reaches the listener 60 with the car navigation display surface 118 as a reflecting surface.

Embodiment 2. FIG.
In this embodiment, stereo signals Ls and Rs based on binaurally recorded sound sources are used as input signals. The sound field generation system 1 according to the second embodiment is configured as shown in FIGS. 19 and 20. Compared to the configuration of the first embodiment, the sound image localization calculation unit 10 is omitted. In other words, the stereo signals Ls and Rs in which the sound image is localized are directly input to the crosstalk cancellation calculation unit 20. In the recording site as shown in FIG. 21, the sound signal from the real sound source P (t) is recorded by the microphones 73 of the left ear 71 and the right ear 72 provided in the dummy head 70. By doing so, stereo signals Ls and Rs to which a transfer function from the real sound source P (t) to the dummy head 70 is added can be obtained. Therefore, the sound image is localized with the position of the real sound source P (t) 74 relative to the dummy head 70 as the virtual sound source p ′ (t). Therefore, regardless of the position of the real sound source P (t) 74, the sound image can be localized for the listener 60 as the virtual sound source p ′ (t) corresponding to the position. As in the first embodiment, the crosstalk cancellation calculation unit 20 changes the transfer function according to the detection result of the sheet sensor 111. Since other processes and installation examples are the same as those in the first embodiment, description thereof is omitted. Even if it is such a structure, the effect similar to said each embodiment can be acquired.

Embodiment 3 FIG.
A sound field generation system 1 according to the present embodiment will be described with reference to FIGS. FIG. 22 is a block diagram showing the configuration of the sound field generation system. FIG. 23 is a top view showing a state in which the sound field generation system is installed in the automobile 100. In the present embodiment, the sound field generation system 1 includes a boarding sensor 112 that detects the presence or absence of a passenger on each seat 110. The crosstalk cancellation calculation unit 20 changes the transfer function according to the detection result of the boarding sensor 112. Specifically, the boarding sensor 112 is provided in the driver's seat 106, the rear seat 107, the passenger seat 108, and the like, and the crosstalk cancellation calculation unit 20 is detected by detecting whether or not the passenger has boarded for each seat. Switches the transfer function according to the layout of the passenger.

  For example, in a state where only the driver is on board and when the driver and the passenger are on board, sound is reflected and absorbed by the passenger, so that the sound field generation environment changes. For example, when a passenger rides on the passenger seat 108, the acoustic signal reflected by the reflecting portion 40 is shielded or reflected by the passenger, so that the passenger is not on the rear seat 107 and the passenger seat 108. The optimal transfer function differs depending on whether only the rear seat 107, only the passenger seat 108, or both the rear seat 107 and the passenger seat 108 are used. In the crosstalk cancellation calculation unit 20, an optimal transfer function corresponding to the presence or absence of a passenger on each seat 110 is set in advance. When the boarding sensor 112 determines which seat 110 the passenger is sitting on, the crosstalk cancellation calculation unit 20 switches the transmission coefficient according to the number of passengers detected by the boarding sensor 112.

  For example, the boarding sensor 112 is a sensor that detects whether or not a seat belt of each seat is attached. When the seat belt provided on each seat is attached, a passenger is seated on the seat. It is determined that the passenger is not seated on the seat in a state where the seat belt is not worn. The boarding sensor 112 may be a force sensor, a contact sensor, or the like that senses that the passenger has got on the seat 110, and the presence or absence of the passenger may be detected for each seat by the force applied by the boarding sensor 112 to the seat. Alternatively, the boarding sensor 112 may be a camera that captures the interior of the automobile 100, and it may be determined whether or not a passenger is sitting on each seat 110 from the interior image captured by the camera. In this way, the boarding sensor 112 detects on which seat of the plurality of seats 110 a person is sitting, and the crosstalk cancellation calculation unit 20 switches the transfer function to the boarding sensor 112 according to the detection result. Thus, the transfer function can be optimized, and a sound field can be generated more appropriately.

Embodiment 4 FIG.
The sound field generation system according to this embodiment will be described with reference to FIG. FIG. 24 is a block diagram showing the configuration of the sound field generation system. The sound field generation system 1 according to the present embodiment includes a window sensor 115 that detects the opening / closing position of the window 114. The crosstalk cancellation calculation unit 20 has the transfer function changing unit 29 shown in FIG. 4 and changes the transfer function according to the detection result of the window sensor 115. As the window 114 in the sound field generation environment opens and closes, the propagation direction of the acoustic signal changes. For example, if there is glass in the window 114 at the position where the acoustic signal is incident, the acoustic signal is reflected. If there is no glass in the window 114, the acoustic signal is not reflected. The field generation environment changes. The crosstalk cancellation calculation unit 20 cancels the crosstalk using an optimal transfer function corresponding to the opening / closing position of the window 114.

  Here, if the right side glass 102 and the left side glass 103 shown in FIGS. 5 and 9 are used as the window 114, the window 114 is opened and closed by moving the right side glass 102 and the left side glass 103 up and down. The sound field generation environment changes. Or if the roof glass 104 is used as the window 114, even if the roof glass 104 moves back and forth, the window 114 opens and closes, and the sound field generation environment changes. A more appropriate sound field can be generated by changing the transfer function according to the opening / closing position of the right side glass 102, the left side glass 103, the roof glass 104, or the like. The transfer function may be changed stepwise or continuously depending on the opening / closing position of the window 114. Furthermore, not only the window 114 but also the transfer function may be changed according to the opening / closing position of a door provided in the sound field generation environment.

  Thus, in Embodiments 1-4, the transfer function is changed according to the sound field generation environment. When the sound field generation environment changes, the transfer characteristics from the left and right speakers to the left and right ears change. Therefore, as in Embodiments 1 to 4, crosstalk can be performed using the optimal transfer function by changing the transfer function. It becomes possible to cancel. Therefore, a sound field can be generated appropriately even in various sound field generation environments. The transfer function corresponding to the sound field generation environment is obtained by impulse response measurement in a sound field generation environment that is substantially the same as the actual sound field generation environment. For example, an impulse response measurement is performed by placing a dummy head in a sound field generation environment that is substantially the same as the actual sound field generation environment. That is, the impulse response is measured by changing the sound field generation environment such as a state where a passenger is on board or a state where the window 114 is opened. By doing so, the transfer functions a11, a12, a21, and a22 shown in FIG. 4 and the like can be optimized.

Embodiment 5 FIG.
The sound field generation system according to the present embodiment will be described with reference to FIGS. FIG. 25 is a block diagram showing the configuration of the sound field generation system, FIGS. 26 and 27 are top views showing the automobile 100 with the sound field generation system 1 installed, and FIG. 28 shows the sound field generation system 1 installed. 1 is a side view showing an automobile 100. FIG. The sound field generation system 1 according to the present embodiment includes an angle changing unit 116 that changes the angle of the speaker unit 30 in addition to the configuration of the fourth embodiment. When the angle changing unit 116 includes a motor for adjusting the angle of the speaker unit 30 and the window 114 is opened from a certain opening / closing position, an acoustic signal from the speaker unit 30 is directly delivered to the listener 60. The angle of the speaker unit 30 is changed so that the speaker unit 30 can be delivered to the listener 60 via the reflecting unit 40 in a closed state from a certain opening / closing position. Other configurations and processes are the same as those in each of the above-described embodiments, and thus description thereof is omitted.

  As shown in FIG. 26, when the right side glass 102 is closed, the acoustic signal from the right speaker 32 is reflected by the right side glass 102 and reaches the right ear 62, and the acoustic signal from the left speaker 31 reaches the left ear 61 directly. ing. As shown in FIG. 27, if the listener 60 sitting on the driver's seat 106 has lowered the right side glass 102 of the window 114, the position where the acoustic signal from the right speaker 32 is incident (see the dotted line arrow in FIG. 27). Since the right side glass 102 is lower than the right side glass 102, the right side glass 102 does not function as the right reflecting portion 42, and an acoustic signal hardly reaches the right ear 62. Therefore, in the present embodiment, the right speaker 32 changes the direction of the sound emission surface so that the sound signal from the right speaker 32 reaches the right ear 62 directly. Or you may make it the left side glass 103 other than the right side glass 102, the windshield 101, or the roof glass 104 grade | etc., Serve as the right reflection part 42. FIG.

  Depending on the position of the right side glass 102 provided in the window 114, the speaker unit 30 changes the direction of the acoustic radiation surface so that the acoustic signal from the speaker unit 30 reaches the listener 60 directly, or the reflection unit 40 Whether to reflect and reach the listener 60 can be selected. In addition, as shown in FIG. 28, when the right side glass 102 falls below the height of the ear (dotted line position in FIG. 28), the speaker unit 30 changes the direction of the sound emitting surface, and the sound signal is received by the listener 60. By directly delivering to the sound field, the sound field can be generated more appropriately even when the sound field generation environment changes. In particular, when the right speaker 32 changes the direction of the sound emission surface according to the open / close position of the side glass closer to the listener 60 (the right side glass 102 in FIGS. 26 to 28), the sound field is generated more appropriately. be able to.

  Note that the direction of the acoustic radiation surface of the right speaker 32 is set in advance for each of the case where the acoustic signal is delivered directly and the case where the acoustic signal is delivered via the right reflector 42, and the right side glass 102 exceeds a predetermined height. The right speaker 32 changes the direction of the acoustic radiation surface. Of course, when the automobile 100 is a left-hand drive vehicle, the direction of the left speaker 31 changes depending on the position of the left side glass 103. The angle changing unit 116 may change the direction of the speaker unit 30 according to the opening / closing positions of the side glasses on both sides as well as the one side glass. Furthermore, as shown in the installation example 8, the crosstalk cancellation calculation unit 20 changes the transfer function according to the angle of the speaker unit 30. When the angle of the right speaker 32 is changed, the crosstalk cancellation calculation unit 20 switches the transfer function, so that an optimal sound field can be generated according to a change in the sound field generation environment.

Embodiment 6 FIG.
A sound field generation device and a sound field generation system according to the present embodiment will be described with reference to FIGS. 29 and 30. FIG. 29 and 30 are top views showing an example in which the sound field generation system 1 is installed in an automobile. Since the sound field generation system 1 according to the present embodiment has the same configuration as that of FIG. 25 of the fifth embodiment, the description of the same contents is omitted.

  In the present embodiment, as shown in FIG. 29, when the roof glass 104 is in the closed state, the roof glass 104 serves as the reflecting portion 40. When the listener 60 opens the roof glass 104 and the roof glass 104 moves behind the position where the acoustic signal reaches (for example, the position indicated by the dotted line in FIG. 30), the roof glass 104 functions as the reflecting portion 40. Therefore, the angle changing unit 116 changes the direction of the speaker unit 30. As a result, as shown in FIG. 30, the acoustic signal from the speaker unit 30 is directly delivered to the listener 60, so that the same effect as in the fifth embodiment can be obtained. Alternatively, the left side glass 103, the front glass 101, the right side glass 102, etc. other than the roof glass 104 may be used as the reflecting portion 40.

  Note that the speaker installation position may be asymmetrical depending on the reflector 40. That is, as viewed from the listener 60, the speakers of each channel do not need to be frontal and symmetrical, and the speaker of the other channel may be placed at a position shifted from front to back and left and right with respect to the speaker of one channel. good. For example, as shown in FIG. 31, the left speaker 31 and the right speaker 32 can be arranged at a left-right asymmetric position. Here, assuming that the center line S in the left-right direction passing through the center of the listener 60 is the center line S, the spread angle θ1 from the center line S to the right speaker 32 is larger than the spread angle θ2 from the center line S to the left speaker 31. It is getting smaller. In other words, the distance to the center line S differs between the left and right speakers. Furthermore, also in the front-rear direction, the distance from the listener 60 to the left speaker 31 is larger than the distance from the listener 60 to the right speaker 32.

  Or as shown in FIG. 32, the position of an up-down direction may differ. Even if the distance from the left ear 61 of the listener 60 to the left speaker 31 and the distance from the right ear 62 of the listener 60 to the right speaker 32 are different in at least one of the vertical direction, the horizontal direction, and the front-back direction. Good.

  In the above description, the sound field generation system 1 is an in-vehicle device, and the sound field generation environment in which the sound field generation system 1 is installed is the inside of the automobile 100. It is not limited to. Further, the reflection unit 40 in the sound field generation system 1 may be other than the vehicle body in the vehicle 100 or the in-vehicle device.

  It is also possible to create a sound field generation environment using a hood such as a motorcycle or a bicycle as a reflection part. Furthermore, it is preferable to mount the sound field generation system 1 in a sound field generation environment that is a closed space. For example, it can be dangerous even in a noisy environment, for example, by installing it in a noisy sound field generation environment such as a driver's cab or operation room in heavy equipment such as an excavator or a game machine in a game hall such as a pachinko parlor. The warning sound can reach the listener's ear from the direction. Furthermore, you may mount in the vehicle for attraction of an amusement park. By doing so, it is possible to generate a more realistic sound field.

  In the above description, the sound field is generated by a 2-channel speaker, but the sound field may be generated by a 3-channel or more speaker. In addition, it is not necessary to reflect the acoustic signals from the speakers of all the channels by the reflecting unit, and the acoustic signals from the speakers of some channels may be directly delivered to the listener 60 without passing through the reflecting unit 40. Good. An acoustic lens that focuses the acoustic signal from the speaker unit 30 on the reflection unit 40 may be provided between the speaker unit 30 and the reflection unit 40. Furthermore, by using a directional speaker for each speaker, reflection to other than the reflection unit 40 can be further prevented.

Embodiment 7.
The sound field generation system 3 according to the seventh embodiment will be described with reference to FIGS. FIG. 33 is a block diagram showing a configuration of the sound field generation system 3 according to the seventh embodiment. FIG. 34 is a diagram schematically illustrating the overall configuration of the sound field generation system 3. Compared to the configuration of the first embodiment, a whispering voice conversion unit 70 is added. The whispering voice conversion unit 70 converts a normal voice into a whispering voice. Specifically, the normal voice, which is a voiced sound, is made unvoiced. As this method, a method of controlling the noise waveform and the amplitude of a periodic sound source waveform disclosed in Japanese Patent Laid-Open No. 58-168098, or Japanese Patent Laid-Open No. 2008- A method for obtaining a transfer function of the vocal tract using the linear prediction method of No. 139573 and multiplying the white noise by using the amplitude ratio with the residual signal subtracted from the original sound waveform, and removing the formant component of JP-A-2006-234969 And other feature conversion methods (eg, A. Kain and MWMacon, “Spectral voice conversion for text-”) for converting feature quantities such as spectral series based on a Gaussian Mixture Model (GMM). to-speechsynthesis "Proc.ICASSP, pp.285-288, Seattle, USA May, 1998.), and any known method can be used.

The voice signal converted into the whisper is input to the sound image localization calculation unit 10 and a convolution calculation is performed. In the present embodiment, since it is provided to the listener's right ear, the transfer function bl (t) of the convolution calculator 11 and the transfer function br (t) of the convolution calculator 12 are the right ear of the real head or dummy head. The transfer function when the sound source is placed on is used, but in order to make the calculation process simpler, the transfer function bl (t) of the convolution calculation unit 11 may be set as shown in the following expression (1).
bl (t) = 0 Formula (1)
The transfer function br (t) of the convolution calculator 12 may be set as shown in the following equation (2).
br (t) = A (t = t ₁ ) A is a constant (amplitude), t ₁ is a predetermined value of t ₁ ≧ 0 br (t) = 0 (t ≠ t ₁ ) (2)

In addition, when providing to a left ear | edge, the value of said transfer function is reverse on right and left.
The procedure after the crosstalk cancellation calculation is the same as in the first embodiment. That is, the crosstalk cancellation calculation unit 20 changes the transfer function according to the position of the seat 110 on which the listener 60 sits, and the crosstalk is canceled with the transfer function according to the relative position between the speaker unit 30 and the listener 60. . Thereby, a transfer function can be optimized and a sound field can be generated more appropriately.

  In this embodiment, the whistling voice is localized and the sound is localized at the ear, so the voice can be heard at the listener's ear, and even if there is a sound that can be heard from the surroundings, such as road noise or car stereo, the voice is clear. Can hear. Also, whispering voices are unvoiced sounds and contain many high frequencies. For those who are not listeners (for example, those who are seated in the front passenger seat if the listener is a driver), it is masked by road noise and car stereo sound because it is far away from the speaker being played. It ’s hard to hear. Therefore, there is an effect that the sound necessary for the listener can be provided only to the listener.

  In this embodiment, the sound image localization calculation is performed after whispering voice conversion, but it goes without saying that the same effect can be obtained even if the audio signals Ls and Rs after the sound image localization calculation are input to the whispering voice conversion unit. For example, when a voice input signal (binaural signal) that has been localized in the ear as a virtual sound source in advance with normal voice is converted into a whispering voice by the whispering voice conversion unit, the whispering voice is provided to the ear and becomes easier to hear.

DESCRIPTION OF SYMBOLS 1 Sound field generation system 2 Sound field generation apparatus 3 Sound field generation system 10 Sound image localization calculation part 11 Convolution calculation part 12 Convolution calculation part 20 Crosstalk cancellation calculation part 29 Transfer function change part 30 Speaker part 31 Left speaker 32 Right speaker DESCRIPTION OF SYMBOLS 40 Reflection part 41 Left reflection part 42 Right reflection part 60 Listener 61 Left ear 62 Right ear 70 Whisper voice conversion part 100 Car 101 Windshield 102 Right side glass 103 Left side glass 104 Roof glass 105 Handle 106 Driver's seat 107 Rear seat 108 Passenger seat 109 Dashboard 110 Seat 111 Seat sensor 112 Boarding sensor 114 Window 115 Window sensor 116 Angle changing unit 117 Car navigation body 118 Car navigation display surface

Claims (14)

A speaker unit that outputs a sound of two or more channels, and that outputs an acoustic signal toward a reflection unit that reflects the sound output from the speaker and transmits the sound to a listener;
An audio signal in which a sound image of an acoustic signal output from the speaker unit and reflected by the reflecting unit to reach the listener is localized as a virtual sound source at the listener's position is input, and the input audio signal On the other hand, a transfer function from the speaker unit to the listener is used so that the crosstalk between the channels of the acoustic signal reflected by the reflecting unit and reaching the listener is canceled at the position of the listener. A crosstalk canceling unit that performs arithmetic processing by
An angle changing unit that changes at least one of the angle of the reflecting surface of the reflecting unit or the angle of the speaker in accordance with a change in a sound field generation environment including the reflecting unit and the listener as components;
A sound field generating device comprising: A speaker unit that outputs a sound of two or more channels, and that outputs an acoustic signal toward a reflection unit that reflects the sound output from the speaker and transmits the sound to a listener;
An audio signal in which a sound image of an acoustic signal output from the speaker unit and reflected by the reflecting unit to reach the listener is localized as a virtual sound source at the listener's position is input, and the input audio signal On the other hand, a transfer function from the speaker unit to the listener is used so that the crosstalk between the channels of the acoustic signal reflected by the reflecting unit and reaching the listener is canceled at the position of the listener. A crosstalk canceling unit that performs arithmetic processing by
A transfer function changing unit that changes the transfer function used in the crosstalk canceling unit in response to a change in a sound field generation environment including the reflecting unit and the listener as components;
A sound field generating device comprising: A sound image localization calculation unit for adding sound image localization so that a sound image of an acoustic signal reaching the listener becomes a virtual sound source at the position of the listener with respect to the audio signal input to the crosstalk canceling unit; And more,
The sound field generation device according to claim 1 or 2, wherein a sound image of an audio signal input to the crosstalk cancellation unit is localized by a calculation by the sound image localization calculation unit. A whispering voice conversion unit for converting a voice signal to be heard by the listener into a whispering voice;
The sound image localization calculation unit outputs a sound image of an acoustic signal output from the speaker unit and reflected by the reflection unit and converted into the whispering voice that has reached the listener as a virtual sound source in at least one ear of the listener The sound field generation device according to claim 3, wherein the calculation is performed so that localization is performed.   5. The transfer function changing unit according to claim 1, wherein the transfer function changing unit changes the transfer function according to a position of a seat on which the listener sits in the sound field generation environment. The sound field generation device described.   The said transfer function change part changes the said transfer function according to the opening / closing position of the window provided in the said sound field generation environment, The Claim 1 characterized by the above-mentioned. Sound field generator.   The transfer function changing unit changes the transfer function according to whether a person other than the listener is sitting on a seat provided in the sound field generation environment. Item 7. The sound field generation device according to any one of items 6 to 6. At least one of the angle of the reflection surface of the reflection part or the angle of the speaker in the speaker part is variable,
The angle of the reflection surface of the front reflection unit or the angle of the speaker is adjusted according to the position of the seat on which the listener is seated in the sound field generation environment. The sound field generation device according to any one of claims 6 to 6.   The sound field generation device according to any one of claims 1 to 8, wherein the input sound signal is a sound signal binaurally recorded at the position of the listener. The sound field generation device according to any one of claims 1 to 8,
A sound field generation system comprising: a reflection unit configured to reflect sound from the speaker unit toward the listener.   The sound field generation system according to claim 10, wherein the reflection unit is an automobile body or an in-vehicle device. The reflection part is at least one of a windshield, a side glass, and a roof glass of an automobile,
The sound field generation system according to claim 10, wherein the transfer coefficient changing unit changes the transfer function according to at least one open / close position of the windshield, the side glass, and the roof glass.   Of the left and right side glasses of the automobile, when the side glass closer to the listener is opened than a predetermined position, the side glass closer to the listener is not used as a reflective part, and the other glass is used as a reflective part. The acoustic signal is delivered to the listener, and when the side glass closer to the listener is closed than a predetermined position, the acoustic signal is transmitted to the listener using the side glass closer to the listener as a reflecting portion. The sound field generation system according to claim 12, wherein an angle of the speaker unit is changed so as to be delivered to the sound field. An audio signal in which a sound image of an acoustic signal output from a speaker unit having a speaker that outputs sound of two or more channels and reflected by a reflection unit and reaches a listener is input as a virtual sound source at the listener's position is input. Steps,
The audio signal is reflected by the reflection unit with respect to the input audio signal and crosstalk between channels of the acoustic signal reaching the listener is canceled from the speaker unit so as to be canceled at the position of the listener. Performing arithmetic processing using a transfer function up to the person,
Outputting an acoustic signal toward a reflecting unit that reflects the sound output from the speaker unit and transmits the sound to a listener;
Changing the transfer function used for arithmetic processing for canceling the crosstalk in response to a change in a sound field generation environment including the reflector and the listener as components;
A sound field generation method comprising:

Images