Acoustic processing system and acoustic processing method

Abstract

To provide an acoustic processing system and an acoustic processing method which are suitable to improvement in the deviation of auditory localization.SOLUTION: In an acoustic processing system including a pre-processing unit and an acoustic processing unit, the acoustic processing unit comprises: an IACF calculation part which acquires a right / left ear mutual correlation function when the sounds output from a plurality of speakers are heard at a prescribed hearing position; a calculation part having a target position decision section which decides a target position on the basis of the right / left ear mutual correlation function within a prescribed range in the right / left ear mutual correlation functions acquired by the IACF calculation part and a delay amount calculation section which calculates a delay amount on the basis of the target position decided by the position decision section; and a processing part which has a delay processing section that delays an audio signal being a signal of the sound output from at least one of the plurality of speakers on the basis of the delay amount calculated by the delay amount calculation section. The right / left ear mutual correlation function within the prescribed range is the right / left ear mutual correlation function within a range of ±n (n is a positive value greater than 1) milliseconds.SELECTED DRAWING: Figure 8

Description

[Technical Field]

[0001] The present invention relates to an acoustic processing system and an acoustic processing method. [Background technology]

[0002] Generally, speakers are installed in multiple locations within a vehicle's interior. For example, the right front speaker on the right door and the left front speaker on the left door are positioned symmetrically across the centerline of the vehicle's interior. However, these speakers are not symmetrically positioned when considered from the listener's listening position (driver's seat, passenger seat, rear seat, etc.).

[0003] For example, when a listener is sitting in the driver's seat, the distance between the right front speaker and the listener is not equal to the distance between the left front speaker and the listener. In the case of a right-hand drive car, for instance, the former distance is shorter than the latter. Therefore, when sound is emitted simultaneously from the speakers in both doors, the listener sitting in the driver's seat will typically hear the sound emitted from the right front speaker first, followed by the sound emitted from the left front speaker. The difference in distance between the listener's listening position and each of the multiple speakers (the difference in the time it takes for the reproduced sound emitted from each speaker to arrive) causes a bias in sound image localization due to the Haas effect.

[0004] Various techniques are known to improve such biases in sound image localization (see, for example, Patent Document 1). [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] Japanese Patent Publication No. 2008-67087 [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] However, the conventional technology exemplified in Patent Document 1 may not be able to sufficiently improve the bias in sound image localization.

[0007] Therefore, in view of the above circumstances, the present invention aims to provide an acoustic processing system and an acoustic processing method suitable for improving bias in sound image localization. [Means for solving the problem]

[0008] An acoustic processing system according to one embodiment of the present invention includes: a function acquisition unit that acquires an interaural cross-correlation function when sound output from multiple speakers is listened to at a predetermined listening position; a position determination unit that determines a target position based on an interaural cross-correlation function within a predetermined range from among the interaural cross-correlation functions acquired by the function acquisition unit; a delay amount calculation unit that calculates a delay amount based on the target position determined by the position determination unit; and a delay unit that delays an audio signal, which is the sound signal, and is output to at least one of the multiple speakers, based on the delay amount calculated by the delay amount calculation unit. The interaural cross-correlation function within a predetermined range is an interaural cross-correlation function in the range of ±n (where n is a positive value greater than 1) milliseconds. [Effects of the Invention]

[0009] According to one embodiment of the present invention, an acoustic processing system and an acoustic processing method suitable for improving bias in sound image localization are provided. [Brief explanation of the drawing]

[0010] [Figure 1] This diagram schematically shows a vehicle equipped with an acoustic processing system according to one embodiment of the present invention. [Figure 2] Block diagram showing the hardware configuration of an acoustic processing device according to one embodiment of the present invention. [Figure 3] This is a functional block diagram of an acoustic processing system according to one embodiment of the present invention. [Figure 4] This is a functional block diagram showing an impulse response acquisition unit related to one embodiment of the present invention. [Figure 5] This is a functional block diagram showing the processing unit related to one embodiment of the present invention. [Figure 6] This flowchart shows the preprocessing performed in the preprocessing unit of one embodiment of the present invention. [Figure 7] This flowchart shows the acoustic processing performed in the acoustic processing unit according to one embodiment of the present invention. [Figure 8] This is a functional block diagram showing a calculation unit related to one embodiment of the present invention. [Figure 9] This figure shows an example of a biaural cross-correlation function calculated by the IACF calculation unit according to one embodiment of the present invention. [Figure 10] This diagram illustrates a method for determining the target position in one embodiment of the present invention. [Figure 11] This figure shows an example of the interaural cross-correlation function calculated by the IACF calculation unit after time alignment processing. [Modes for carrying out the invention]

[0011] The following description relates to an acoustic processing system and an acoustic processing method according to one embodiment of the present invention.

[0012] Figure 1 is a schematic diagram showing a vehicle A (for example, a right-hand drive vehicle) equipped with an acoustic processing system 1 according to one embodiment of the present invention. As shown in Figure 1, the acoustic processing system 1 consists of an acoustic processing device 2 and a pair of left and right speakers SP FR SP FL It also features a binaural microphone.

[0013] Speaker SP FR This is the right front speaker embedded in the right door (driver's side door). FL This is the left front speaker embedded in the left door (passenger side door). Vehicle A may also have other speakers installed (for example, rear speakers) (i.e., three or more speakers installed).

[0014] A binaural microphone (MIC) is configured, for example, by incorporating microphones into each ear of a dummy head that mimics a human head. The microphone incorporated into the right ear of the dummy head is referred to as the "microphone MIC." R It is written as follows: The microphone built into the left ear of the dummy head is labeled "Microphone MIC L It is written as follows:

[0015] Figure 2 is a block diagram showing the hardware configuration of the sound processing device 2. As shown in Figure 2, the sound processing device 2 comprises a player 10, an LSI (Large Scale Integration) 11, a D / A converter 12, an amplifier 13, a display unit 14, an operation unit 15, and a flash memory 16.

[0016] Player 10 is connected to the sound source. Player 10 plays the audio signal input from the sound source and outputs it to LSI 11.

[0017] The sound source is, for example, a disc media such as a CD (Compact Disc) or SACD (Super Audio CD) containing digital audio data, or a storage media such as an HDD (Hard Disk Drive) or USB (Universal Serial Bus). A telephone (e.g., a feature phone or smartphone) may also be the sound source. In this case, the player 10 outputs the voice signal from the telephone during a call to the LSI 11.

[0018] LSI11 is an example of a computer equipped with a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), etc. The CPU of LSI11 includes a single processor or a multiprocessor (in other words, at least one processor) that executes programs written to the ROM of LSI11, and comprehensively controls the sound processing device 2.

[0019] LSI11 obtains the interaural cross correlation function (IACF) of the sound output from a plurality of speakers (in this embodiment, speakers SP FR , SP FL ) when listened to at a predetermined listening position (e.g., the driver's seat, the passenger seat, or the rear seat), determines a target position based on the interaural cross correlation functions within a predetermined range among the obtained interaural cross correlation functions, calculates a delay amount based on the determined target position, and delays an audio signal, which is a sound signal, output to at least one of the plurality of speakers based on the calculated delay amount. The interaural cross correlation functions within the predetermined range are the interaural cross correlation functions within a range of ±n (where n is a positive value greater than 1) milliseconds (msec).

[0020] The audio signal after the above time alignment processing by LSI11 is converted into an analog signal by D / A converter 12. This analog signal is amplified by amplifier 13 and output to speakers SP FR , SP FL . Thereby, for example, a music piece recorded in a sound source is reproduced in the vehicle interior from speakers SP FR , SP FL .

[0021] According to this embodiment, by calculating the delay amount using the interaural cross correlation functions within a wide range exceeding the range of ±1 millisecond (i.e., the range of ±n milliseconds) and performing time alignment processing, the deviation of sound image localization that is likely to occur in a listening environment such as the vehicle interior is improved.

[0022] In this embodiment, an in-vehicle acoustic processing system 1 is exemplified. However, the deviation of sound image localization may also occur in a listening environment such as the interior of a building. Therefore, the acoustic processing system 1 may be implemented for a listening environment other than the vehicle interior.

[0023] The display unit 14 is a device that displays various screens, including setting screens, and includes displays such as LCDs (Liquid Crystal Displays) and organic ELs (Electro Luminescence). The display unit 14 may also include a touch panel.

[0024] The operation unit 15 includes switches, buttons, knobs, wheels, and other operating elements of the mechanical, capacitive non-contact, and membrane types. If the display unit 14 includes a touch panel, this touch panel also forms part of the operation unit 15.

[0025] Figure 3 is a functional block diagram of the sound processing system 1. The functions shown in each block in Figure 3 and other functional block diagrams are executed through the cooperation of the software and hardware provided in the sound processing system 1.

[0026] As shown in Figure 3, the acoustic processing system 1 includes a pre-processing unit 100 and an acoustic processing unit 200 as functional blocks.

[0027] The pre-processing unit 100 performs pre-processing to improve bias in sound image localization. As shown in Figure 3, the pre-processing unit 100 includes an impulse response acquisition unit 101 and an impulse response recording unit 102.

[0028] Figure 4 is a functional block diagram showing the impulse response acquisition unit 101. As shown in Figure 4, the impulse response acquisition unit 101 includes a measurement signal generation unit 101a, a control unit 101b, and a response processing unit 101c as functional blocks.

[0029] The measurement signal generation unit 101a generates a predetermined measurement signal. The generated measurement signal is, for example, an M-sequence code (Maximal length sequence). The length of this measurement signal is set to at least twice the code length. Note that the measurement signal may also be of other types, such as a TSP signal (Time Stretched Pulse).

[0030] The control unit 101b receives the measurement signal from the measurement signal generation unit 101a and outputs it to each speaker SP. FR SP FL The sound is output sequentially to each speaker SP. This allows a predetermined measurement sound to be output at predetermined time intervals. FR SP FL The output will be generated sequentially from there.

[0031] In this embodiment, the measurement position for the impulse response (an example of a predetermined listening position) is the driver's seat. Therefore, the binaural microphone (MIC) is installed in the driver's seat. The installation position of the binaural microphone (MIC) changes depending on the listening position.

[0032] Microphone MIC R and microphone MIC L First, the speaker SP FR The measurement sound output from the microphone is captured. R and microphone MIC L Next, speaker SP FL The measurement sound output from the device is collected.

[0033] The control unit 101b controls the microphone MIC R MIC L The measurement sound signal (i.e., measurement signal) collected by each of these is output to the response processing unit 101c. Below, speaker SP FR Output from the microphone MIC R The measurement signal captured is "Measurement signal R R It is written as "Speaker SP". FL Output from the microphone MIC R The measurement signal captured is "Measurement signal R L It is written as "Speaker SP". FR Output from the microphone MIC L The measurement signal captured is "Measurement signal L R It is written as "Speaker SP". FL Output from the microphone MIC L The measurement signal captured is "Measurement signal L L It is written as follows:

[0034] The response processing unit 101c acquires the impulse response.

[0035] For example, the response processing unit 101c processes the measurement signal R R The cross-correlation function between the measurement signal and the reference signal is calculated to determine the impulse response, and the measurement signal R L The cross-correlation function between the measurement signal and the reference signal is calculated to determine the impulse response, and the two calculated impulse responses are combined. The combined impulse response corresponds to the listener's right ear. Hereafter, the impulse response corresponding to the listener's right ear will be denoted as "Impulse Response R'".

[0036] The response processing unit 101c receives the measurement signal L R The cross-correlation function between the measurement signal and the reference signal is calculated to determine the impulse response, and the measurement signal L L The cross-correlation function between the measurement signal and the reference signal is calculated to determine the impulse response, and the two calculated impulse responses are combined. The combined impulse response corresponds to the listener's left ear. Hereafter, the impulse response corresponding to the listener's left ear will be denoted as "Impulse Response L'".

[0037] The reference measurement signal is identical to and time-synchronized with the measurement signal generated by the measurement signal generation unit 101a. The reference measurement signal is stored, for example, in the flash memory 16.

[0038] The impulse response recording unit 102 writes the impulse responses R' and L' acquired by the impulse response acquisition unit 101 to, for example, the flash memory 16.

[0039] As shown in Figure 3, the acoustic processing unit 200 includes a band division unit 201, a calculation unit 202, an input unit 203, a band division unit 204, a processing unit 205, a band synthesis unit 206, and an output unit 207.

[0040] The band division unit 201 includes, for example, a 1 / N octave band filter. The band division unit 201 divides the impulse responses R' and L' written to the flash memory 16 into multiple bands bw1 to bwN using the 1 / N octave band filter and outputs them to the calculation unit 202.

[0041] Hereafter, the impulse response R' of each divided band will be denoted as "divided band response Rd," and the impulse response L' of each divided band will be denoted as "divided band response Ld."

[0042] The calculation unit 202 generates various control parameters by performing the following processes for each bandwidth bw1 to bwN: calculation of the interaural cross-correlation function based on the divided bandwidth response Rd and divided bandwidth response Ld, determination of the target position based on the calculated interaural cross-correlation function, calculation of the delay amount based on the target position, and calculation of the phase correction amount. Details of each process performed by the calculation unit 202 will be described later.

[0043] The various control parameters generated by the calculation unit 202 include control parameters CPd and CPp corresponding to each of the bandwidths bw1 to bwN. The control parameter CPd is for speaker SP FR The audio signal output to the speaker SP FL This is a control parameter for delaying one of the audio signals output to the device. The control parameter CPp is a control parameter for determining the amount of phase correction of the audio signal by the all-pass filter.

[0044] The input unit 203 includes a selector connected to various sound sources. The input unit 203 outputs the audio signal S1 input from the sound source connected to the selector to the band division unit 204.

[0045] In this embodiment, the audio signal S1 is the audio signal S1 of the R channel. R and the audio signal S1 of the left channel L It is assumed to be a 2-channel signal including [the specified element].

[0046] The band division unit 204 includes, for example, a 1 / N octave band filter. The band division unit 204 divides the audio signal S1 input from the input unit 203 into multiple bands bw1 to bwN using a 1 / N octave band filter, similar to the band division unit 201, and outputs them to the processing unit 205.

[0047] Below are the audio signals S1 of each band after division. R This is "Divided-band audio signal S2 R It is written as follows: Also, the audio signals S1 of each band after division L This is "Divided-band audio signal S2 L It is written as follows:

[0048] Figure 5 is a functional block diagram showing the processing unit 205. As shown in Figure 5, the processing unit 205 includes a delay processing unit 205a and a phase correction unit 205b.

[0049] The delay processing unit 205a delays the audio signal for each bandwidth bw1 to bwN. For example, for each bandwidth bw1 to bwN, the delay processing unit 205a delays the divided bandwidth audio signal S2 input from the bandwidth division unit 204 based on the control parameter CPd input from the calculation unit 202. R and divided-band audio signal S2 L One of the signals is delayed and output to the phase correction unit 205b.

[0050] The phase correction unit 205b corrects the phase of the audio signal for each bandwidth bw1 to bwN. For example, the phase correction unit 205b includes an all-pass filter. As will be described in more detail later, if the sign of the correlation value of the biaural cross-correlation function is negative, the phase correction unit 205b corrects the divided bandwidth audio signal S2 based on the control parameter CPp input from the calculation unit 202. R and S2 L An all-pass filter is applied to correct the phase, and the signal is output to the band synthesis unit 206. Also, if the sign of the correlation value of the biaural cross-correlation function is positive, the phase correction unit 205b processes the divided band audio signal S2 R and S2 LThe output is sent to the band blending unit 206 without applying an all-pass filter.

[0051] The following is the divided-band audio signal S2 output from the phase correction unit 205b. R This is "Divided-band audio signal S3 R It is written as follows: In addition, the divided-band audio signal S3 is output from the phase correction unit 205b. L This is "Divided-band audio signal S3 L It is written as follows:

[0052] The band combination unit 206 receives divided band audio signals S3 of bands bw1 to bwN from the phase correction unit 205b. R The divided-band audio signals S3 of bands bw1 to bwN input from the phase correction unit 205b are combined, and the divided-band audio signals S3 L The audio signals S3 are synthesized from divided bandwidths bw1 to bwN. R The combined audio signal S4 for the R channel R And, divided bandwidth audio signal S3 of bandwidth bw1~bwN L The L channel audio signal S4 is a composite of these. L This is output to output unit 207.

[0053] The output unit 207 receives the two-channel audio signal S4 input from the bandwidth combining unit 206. R S4 L These are each converted into analog signals, and the converted analog signals are amplified to power the speaker SP FR SP FL The signal is output into the vehicle cabin. This allows, for example, the playback of a musical track. In the delay processing unit 205a, time alignment processing based on the control parameter CPd is performed, thereby improving the bias in sound image localization during music playback.

[0054] Figure 6 is a flowchart showing the preprocessing performed by the preprocessing unit 100 according to one embodiment of the present invention. For example, when a predetermined touch operation is performed on the display unit 14 or a predetermined operation is performed on the operation unit 15, the execution of the preprocessing shown in Figure 6 is started. In order to perform the preprocessing, a binaural microphone MIC is installed at the listening position (for example, the driver's seat).

[0055] In the preprocessing shown in Figure 6, the measurement signal generation unit 101a generates a predetermined measurement signal (step S101). The control unit 101b then sends this measurement signal to each speaker SP FR SP FL Output sequentially (step S102).

[0056] Binaural microphones are used with each speaker. FR SP FL The measurement sounds output sequentially from there are collected (step S103).

[0057] The control unit 101b receives the measurement signal (specifically, the measurement signal R) from the binaural microphone MIC. R , R L , L R and L L ) is output to the response processing unit 101c.

[0058] The response processing unit 101c receives the measurement signal R from the control unit 101b. R and R L Based on this, the impulse response R' is calculated, and the measurement signal L input from the control unit 101b is also used. R and L L The impulse response L' is calculated based on this (step S104). The impulse response recording unit 102 writes the impulse responses R' and L' calculated by the response processing unit 101c to the flash memory 16 (step S105).

[0059] Figure 7 is a flowchart showing the acoustic processing performed by the acoustic processing unit 200 according to one embodiment of the present invention. For example, when the impulse response recording unit 102 writes the impulse responses R' and L' to the flash memory 16, the execution of the acoustic processing shown in Figure 7 begins.

[0060] In the acoustic processing shown in Figure 7, the band division unit 201 divides the impulse responses R' and L' written to the flash memory 16 into multiple bands bw1 to bwN (step S201). The divided band responses Rd and Ld of each band after division are input to the calculation unit 202.

[0061] Figure 8 is a functional block diagram showing the calculation unit 202. As shown in Figure 8, the calculation unit 202 includes an IACF calculation unit 202a, a target position determination unit 202b, a delay amount calculation unit 202c, and a phase correction amount calculation unit 202d.

[0062] The IACF calculation unit 202a calculates the interauricular cross-correlation function for each bandwidth bw1 to bwN (step S202). For example, the IACF calculation unit 202a calculates the interauricular cross-correlation function using the following formula.

[0063] (formula) TIFF0007874455000001.tif21162

[0064] Rd(t) represents the amplitude of the divided-band response Rd at time t, and indicates the sound pressure entering the right ear at time t. Ld(t) represents the amplitude of the divided-band response Ld in the same band as the above divided-band response Rd at time t, and indicates the sound pressure entering the left ear at time t. t1 and t2 indicate the measurement time. For example, t1 is 0 milliseconds and t2 is 100 milliseconds. τ represents the correlation time. The range of the correlation time τ is greater than ±1 millisecond, and is typically within the range of ±50 milliseconds.

[0065] Figure 9 shows the interauricular cross-correlation function calculated by the IACF calculation unit 202a. As an example, Figure 9 shows the interauricular cross-correlation function for one of the bands bw1 to bwN. In Figure 9, the vertical axis represents the correlation value, and the horizontal axis represents the correlation time (unit: msec).

[0066] The closer the sound waveforms reaching the listener's right and left ears are, the closer the absolute value of the correlation in the biaural cross-correlation function, as illustrated in Figure 9, approaches 1. If the sounds reaching the listener's right and left ears are in the same phase, the correlation value is positive; if the sounds reaching the listener's right and left ears are in opposite phases, the correlation value is negative. A higher absolute value of the correlation indicates a stronger sense of sound localization, while a lower absolute value indicates a weaker sense of sound localization.

[0067] In this embodiment, the correlation value is calculated based on the right ear. Therefore, when the sound image is located to the right of the listener, a high peak correlation value is likely to appear during positive times. Conversely, when the sound image is located to the left of the listener, a high peak correlation value is likely to appear during negative times. From this, it can be inferred that in the example shown in Figure 9, the sound image is localized slightly to the right of the listener.

[0068] Thus, the IACF calculation unit 202a calculates multiple speakers (speaker SP) FR SP FL It operates as a function acquisition unit that obtains the binaural cross-correlation when the sound output from ) is listened to at a predetermined listening position (e.g., driver's seat, passenger seat, or rear seat).

[0069] In this embodiment, the following process is performed to improve the slightly rightward bias in sound image localization, as illustrated in Figure 9.

[0070] For example, the target position determination unit 202b determines the target position for each band bw1 to bwN based on the biaural cross-correlation function calculated in step S202 (step S203).

[0071] Figure 10 is a diagram in which symbols and other information have been added to Figure 9 to explain the method for determining the target position. The target position determination unit 202b calculates the centroid C of the biaural cross-correlation function within a predetermined range on a coordinate plane, as illustrated in Figure 9, where the vertical axis is the correlation value and the horizontal axis is time.

[0072] A predetermined range of interauricular cross-correlation functions is, for example, an interauricular cross-correlation function within a range of ±30 milliseconds. The centroid C is the centroid of the entire figure formed by the interauricular cross-correlation functions within a range of ±30 milliseconds on the coordinate plane. The figure formed by the interauricular cross-correlation functions is the figure shown by the hatched region (see Figure 10) enclosed by the line with a correlation value of 0 and the graph of the interauricular cross-correlation function.

[0073] The target position determination unit 202b determines the calculated centroid C as the target position.

[0074] In another embodiment, the target position determination unit 202b may determine the peak position of the biaural cross-correlation function near the centroid C as the target position. Illustratively, the target position determination unit 202b may determine the peak position P1 closest to the centroid C as the target position, or it may determine the largest peak position P2 within a certain range (for example, a range of ±10 milliseconds centered on the centroid C) as the target position.

[0075] Thus, the target position determination unit 202b operates as a position determination unit that determines the target position based on the interauricular cross-correlation function within a predetermined range (±n milliseconds) from the interauricular cross-correlation function obtained by the IACF calculation unit 202a. Furthermore, the target position determination unit 202b operates as a centroid calculation unit that calculates the centroid C of the interauricular cross-correlation function within a predetermined range on a coordinate plane with the vertical axis representing the correlation value and the horizontal axis representing time, and determines the target position based on this centroid.

[0076] The delay amount calculation unit 202c calculates the delay amount for each bandwidth bw1 to bwN based on the target position determined by the target position determination unit 202b (step S204).

[0077] Exemplarily, the delay amount calculation unit 202c calculates the delay amount for the audio signal output to one speaker SP so that the center of gravity C, which is the target position, is located at 0 seconds or near 0 seconds on the time axis. In the present embodiment, since the center of gravity C appears at a position where it is T C seconds on the time axis (in other words, slightly to the right of the listener), the delay amount calculation unit 202c calculates T FR seconds as the delay amount for the audio signal output to the speaker SP. C seconds.

[0078] The delay amount calculation unit 202c generates a control parameter CPd for delaying the audio signal to be delayed for each of the bands bw1 to bwN (step S205).

[0079] The control parameter CPd includes values indicating the delay target and its delay amount. In the examples of FIGS. 9 and 10, the value indicating the audio signal output to the speaker SP as the delay target and the value indicating T FR seconds as the delay amount are included in the control parameter CPd. C seconds.

[0080] When the target position is the peak position P1, the delay amount calculation unit 202c calculates T FR seconds as the delay amount for the audio signal output to the speaker SP. When the target position is the peak position P2, the delay amount calculation unit 202c calculates T P1 seconds as the delay amount for the audio signal output to the speaker SP. FR seconds as the delay amount for the audio signal output to the speaker SP. P2 seconds.

[0081] The acoustic processing unit 200 executes time alignment processing based on the control parameter CPd (step S206).

[0082] Specifically, the delay processing unit 205a of the processing unit 205 performs delay processing based on the control parameter CPd for each bandwidth bw1 to bwN. Next, bandwidth combining processing is performed by the bandwidth combining unit 206 and output processing is performed by the output unit 207, and audio signals that have undergone time alignment processing for each bandwidth bw1 to bwN are reproduced.

[0083] Thus, the delay processing unit 205a operates as a delay unit that delays the audio signal output to at least one of the multiple speakers based on the delay amount calculated by the delay amount calculation unit 202c.

[0084] In the preprocessing unit 100, the impulse responses R' and L' of the sound after time alignment processing, which are output from the output unit 207, are calculated and written to the flash memory 16 (see steps S103 to S106 in Figure 6).

[0085] The band division unit 201 divides the impulse responses R' and L' of the sound after time alignment processing, which have been written to the flash memory 16, into multiple bands bw1 to bwN (step S207). The IACF calculation unit 202a calculates the interaural cross-correlation function of the impulse responses R' and L' of the sound after time alignment processing for each band bw1 to bwN (step S208).

[0086] Figure 11 shows an example of the interaural cross-correlation function calculated by the IACF calculation unit 202a in step S208.

[0087] As shown in Figure 11, time alignment processing based on the control parameter CPd is performed, causing the centroid C of the biaural cross-correlation function within a predetermined range (±30 milliseconds) to move to a position near 0 seconds on the time axis. In the example in Figure 11, it can be seen that the bias in sound image localization is improved because the centroid C, which corresponds to a strong sense of sound image localization, is located near 0 seconds on the time axis.

[0088] In this embodiment, the target position is determined not by a simple method such as determining the highest peak position as the target position, but by the center of gravity, which also takes into account correlation values ​​other than the peak position (in other words, values ​​that affect the localization of the sound image). Therefore, even in listening environments such as car interiors, where the speaker arrangement is asymmetrical and there are many reflected and reverberating sounds, the graph of the binaural cross-correlation function can take on a complex shape, and a sufficient effect of improving the bias in sound image localization can be obtained.

[0089] Here, if the sign of the correlation value with the largest absolute value among the biaural cross-correlation functions within a predetermined range calculated in step S208 is negative, then at a position where the sense of sound localization is strong, speaker SP FR Sound from the speaker SP FL The sound is in phase with the source sound. Therefore, the listener perceives this as an audible distortion.

[0090] Therefore, if the sign of the largest correlation value is negative (step S209: YES), the phase correction amount calculation unit 202d generates a control parameter CPp to make the sign of this correlation value positive (step S210). If the sign of the largest correlation value is positive (step S209: NO), the acoustic processing shown in Figure 7 is completed.

[0091] The control parameter CPp includes a value indicating the phase correction amount. The phase correction amount, for example, indicates a value that rotates the phase of the processing band by 180° among the bands bw1 to bwN.

[0092] The acoustic processing unit 200 performs phase correction processing based on the control parameter CPp (step S211).

[0093] Specifically, the phase correction unit 205b of the processing unit 205 performs phase correction processing based on the control parameter CPp for each of the frequency bands bw1 to bwN using an all-pass filter. The all-pass filter applied in the phase correction processing is, for example, a cascade connection of a predetermined number of second-order IIR (Infinite Impulse Response) filters. Note that the number of second-order IIR filters is appropriately determined in consideration of the phase correction accuracy and the filter processing load.

[0094] Due to the phase correction processing by the phase correction unit 205b, the phase of the sound from the speaker SP FR and the sound from the speaker SP FL becomes aligned, so that music or the like is reproduced as a natural sound in terms of the sense of hearing.

[0095] The above is the description of the exemplary embodiments of the present invention. The embodiments of the present invention are not limited to those described above, and various modifications are possible within the scope of the technical idea of the present invention. For example, the content obtained by appropriately combining the embodiments explicitly exemplified in the specification or obvious embodiments is also included in the embodiments of the present application.

[0096] For example, in the above embodiment, the calculation and recording of the impulse responses R' and L' are performed as preprocessing for improving the bias in sound image localization, but the present invention is not limited to this. In another embodiment, in addition to the calculation and recording of the impulse responses R' and L', band division by the band division unit 201 and various processes by the calculation unit 202 (calculation of the interaural cross-correlation function, determination of the target position, calculation of the delay amount, calculation of the phase correction amount, generation of the control parameter) may be performed as preprocessing.

[0097] The speaker SP FR and SP FL In addition, when a pair of speakers is also installed on the rear seat side, the processing is executed according to the following procedure. Exemplarily, a binaural microphone MIC is installed in the front seat (driver's seat or passenger seat), and the speakers SP FR and SP FLThe processes shown in Figures 6 and 7 are performed on the target. Next, a binaural microphone MIC is installed in the rear seat, and the processes shown in Figures 6 and 7 are performed on the pair of speakers on the rear seat side. [Explanation of symbols]

[0098] 1: Acoustic processing system 2: Acoustic processing device 100: Pre-processing 200: Acoustic Processing Unit

Claims

1. A function acquisition unit that acquires the binaural cross-correlation function when sound output from multiple speakers is listened to at a predetermined listening position, A position determination unit determines the target position based on interauricular cross-correlation functions within a predetermined range, among the interauricular cross-correlation functions obtained by the function acquisition unit. A delay amount calculation unit calculates a delay amount based on the target position determined by the position determination unit, A delay unit delays the audio signal, which is the sound signal, that is output to at least one of the plurality of speakers, based on the delay amount calculated by the delay amount calculation unit, The system includes a centroid calculation unit that calculates the centroid of the interaural cross-correlation function within a predetermined range on a coordinate plane where the vertical axis represents the correlation value and the horizontal axis represents time, The position determination unit determines the target position based on the centroid of the biaural cross-correlation function calculated by the centroid calculation unit. Acoustic processing system.

2. The target position is the centroid of the interaural cross-correlation function within the predetermined range, or the peak position of the interaural cross-correlation function near the centroid. The acoustic processing system according to claim 1.

3. If the sign of the correlation value that forms the peak of the biaural cross-correlation function after the delay processing of the audio signal by the delay unit is negative, the phase of the audio signal is corrected so that the sign of the correlation value becomes positive. The acoustic processing system according to claim 1 or claim 2.

4. The function acquisition unit acquires the interaural cross-correlation function corresponding to each of the multiple frequency bands, For each of the aforementioned multiple bandwidths, the position determination unit determines the target position, the delay amount calculation unit calculates the delay amount, and the delay unit performs delay processing on the audio signal. The acoustic processing system according to claim 1 or claim 2.

5. The interaural cross-correlation function is obtained when sounds output from multiple speakers are listened to at a predetermined listening position. On a coordinate plane with the vertical axis representing correlation value and the horizontal axis representing time, the centroid of the biaural cross-correlation function within a predetermined range is calculated. Based on the centroid of the interauricular cross-correlation function within the predetermined range calculated from the acquired interauricular cross-correlation function, the target position is determined. Based on the determined target position, the delay amount is calculated. Based on the calculated delay amount, the computer is instructed to perform a process that delays the audio signal, which is the sound signal, and is output to at least one of the multiple speakers. Sound processing methods.

Images