JP2001236077A - Delay time setting system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a delay time setting system by which sound quality at a listening position can be improved when a controlled sound source and an uncontrolled sound source are used at the same time. SOLUTION: A prescribed signal for measuring a delay time is outputted from a sound source 50 for measuring the delay time, and the corresponding sound waves are radiated in an acoustic space in a car room from a loudspeaker 32. When the prescribed signal is outputted from the sound source 50 for measuring a delay time and a corresponding detection signal is outputted from a microphone 14, a delay time calculating/setting part 52 obtains a frequency characteristic and a grouped delay time characteristic of sound pressure levels based on the detected signal. Next, the delay time calculating/setting part 52 obtains a delay time t1 at a frequency f1 at the boundary position between a controlled frequency band and an uncontrolled frequency band based on the grouped delay time characteristic, and calculates the difference τ=T1-t1 between a known delay time value T1 from the loudspeaker 20 up to the microphone 14 and the delay time t1, and sets a delaying unit 26 to this difference τ as the delay time thereof.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of setting a delay time of a non-controlled sound source in an equalization system for realizing desired acoustic characteristics in a vehicle interior or the like by using both a controlled sound source and a non-controlled sound source.

[0002]

2. Description of the Related Art Since a vehicle interior is a closed narrow space, reflection occurs in a short time, and sound waves interfere with each other, so that the transmission characteristics to a listening position become very complicated. In addition, since the user listens to music in an asymmetrical location, the transfer characteristics from the left and right speakers are also significantly different. There is a demand for an audio device for removing such adverse effects in the vehicle interior and improving the acoustic characteristics in the vehicle interior. To meet such demands, the transmission characteristics of the acoustic system are automatically corrected. A known equalization system is known.

FIG. 8 is a diagram showing a basic configuration of an equalization system applied to an in-vehicle audio system. As an example, a configuration for correcting a transfer characteristic in a time domain is shown. The equalization system shown in FIG. 1 includes a target response setting unit 100, a microphone 102, an adding unit 104, a speaker 108, an LMS (Least Mean Square) algorithm processing unit 110, an adaptive filter 112, and a filter 114.

In the target response setting section 100, a transfer characteristic corresponding to an acoustic space to be reproduced is set as a target response characteristic H. For example, flat characteristics are set in all audio frequency bands in which an input signal is simply delayed for a predetermined time. The adaptive filter 112 has a FIR (Fini
A digital filter of te Impulse Response) type is used, and a filter coefficient vector W (n) is set by the LMS algorithm processing unit 110. The LMS algorithm processing unit 110 calculates an error signal that is a difference between the audio signal collected and output by the microphone 102 installed at the listening position of the listener and the target response signal output from the target response setting unit 100. In order to minimize the power, the adaptive filter 112 using the LMS algorithm is used.
Is set as the filter coefficient vector W (n). The filter 114 has a characteristic obtained by modeling a transfer characteristic of a vehicle interior acoustic system from the speaker 108 to the microphone 102. With such an equalization system, it is possible to listen to the same music as when listening to music in a space having the target response characteristic H set in the target response setting unit 100.

[0005] By the way, if the equalization processing is performed on the entire band of the audio signal as in the above-described equalization system, the amount of calculation becomes enormous, and real-time processing becomes difficult. For example, if processing is to be performed in real time, several hundred DSPs (digital signal processors) are required. Therefore, there has been proposed an equalization system that performs processing only in a specific frequency band, for example, only in a low frequency range of 200 Hz or less.

FIG. 9 is a diagram showing a configuration of an equalization system that performs an adaptive equalization process only in a low frequency range. In the equalization system shown in the figure, a low-pass filter (LPF) 116 that allows a low-frequency range to pass is inserted into the input side of the equalization system shown in FIG. The point input to 112 etc.
A high-pass filter (HP
F) The second speaker 1 that inserts 120 and emits high-range audio sound to the vehicle interior acoustic space based on the output of the 120
The second embodiment is different from the second embodiment in that a delay unit 124 for delaying an audio signal by a predetermined time is inserted in a stage preceding the second speaker 122.

In the equalization system shown in FIG. 9, performing the equalization processing only on the low frequency range of the audio signal has the following advantages. The sampling frequency, that is, the required amount of calculation can be reduced. In a small space such as a vehicle interior, in the low-frequency range, the mode overlap is small and the reflectance of the wall is high, so that the effect of the standing wave is remarkable. Further, speakers used in in-vehicle audio devices are limited in size and mounting state, and it is difficult to efficiently reproduce sound in a low sound range. For this reason, by concentrating the control on the bass range, a large control effect can be expected for a small hardware scale.

[0008]

By the way, in the above-described equalization system for performing equalization processing on a bass range,
Since a predetermined delay is caused by passing the sound in the low-frequency range that is the control band through the adaptive filter 112, in order to adjust this delay time, the delay unit 124 is provided in front of the speaker 122 that is the uncontrolled sound source as described above. Is provided. FIG. 10 is a diagram illustrating a conventional method of setting a delay time. FIG. 10A shows frequency characteristics of sound pressure levels of sound waves corresponding to the control band (low sound range) and the non-control band (high sound range), respectively. FIG. 6 shows group delay characteristics (frequency characteristics of delay time) of sound waves corresponding to respective control bands. Conventionally, as shown in FIG. 10B, an average delay time until a sound wave corresponding to a non-control band reaches the microphone 102 installed at the listening position is measured, and the average delay time is measured. The delay time of the delay unit 124 is set so as to be substantially the same between the control band and the non-control band.

However, as in the example shown in FIG. 10B, the delay time is set by focusing on the frequency f ₀ at the boundary position between the control band and the non-control band (hereinafter referred to as “boundary frequency”). As seen, the delay time T ₀ corresponding to the sound wave in the control band and the delay time t ₀ corresponding to the sound wave in the non-control band.
Often does not match. As described above, in the vicinity of the boundary frequency f ₀ , if the delay time T ₀ corresponding to the sound wave in the control band and the delay time t ₀ corresponding to the sound wave in the non-control band do not match, the sound wave having the same frequency is generated. The timing at which both the sound wave in the control band and the sound wave in the non-control band arrive at the listening position is shifted, which causes a problem of a sense of incongruity and deteriorates sound quality.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to provide a delay time capable of improving sound quality at a listening position when a control sound source and a non-control sound source are used together. To provide a setting method.

[0011]

In order to solve the above-mentioned problems, according to the delay time setting method of the present invention, a sound collecting means is set at a listening position, and a control sound source and a non-control sound source are installed in an acoustic space. A delay means capable of setting a delay time is connected to a stage preceding the non-controlled sound source. A predetermined delay time measurement signal is output from the measurement signal generation means, and a sound wave corresponding to this signal is emitted from the uncontrolled sound source. Then, by analyzing the signal output from the sound collecting means by the delay time calculating / setting means, the overlapping portion of the frequency included in the sound waves radiated from each of the uncontrolled sound source and the control sound source reaches the sound collecting means. The delay time of the delay means is set so that the timings substantially match. Since the delay time of the delay means is set so that the overlapping portions of the sound waves radiated from the control sound source and the non-control sound source arrive at the sound collection means substantially coincide with each other, the listening position is the same frequency of the sound waves. Can be prevented from causing a sense of incongruity due to a shift in the timing of reaching, and the sound quality at the listening position can be improved.

The delay time calculating / setting means analyzes the signal output from the sound collecting means in response to the sound wave of the delay time measuring signal emitted from the uncontrolled sound source,
It is preferable to calculate a delay time corresponding to the frequency of the overlapping portion, and set the delay time of the delay unit so that the calculated delay time matches a known delay time corresponding to the control sound source. Since the delay time of the delay means is set so that there is no time difference between the emission of the sound wave from each of the uncontrolled sound source and the control sound source and the arrival at the sound collection means, the overlapping portion of the frequency contained in each sound wave is collected. The timing of reaching the sound means can be surely matched.

A time-stretched pulse is used as the delay time measuring signal, and a signal obtained by inverting the time-stretched pulse on the time axis is added to the output signal of the sound collecting means by the delay time calculating / setting means. However, it is desirable to obtain a delay time at the frequency of the overlapping portion corresponding to the uncontrolled sound source by performing a Fourier transform process on the convolution operation result. By performing such a convolution operation, an impulse response can be obtained, and a delay time corresponding to a frequency overlapping portion can be obtained based on the impulse response. The time-stretched pulse is a signal having a predetermined time width and a frequency component dispersed on the time axis, and has an advantage that it is hardly affected by sudden noise.

Further, a white noise signal may be used as the delay time measuring signal. In this case,
The delay time calculation / setting means receives the delay time measurement signal output from the measurement signal generation means and the output signal of the sound collection means, and minimizes the power of the error signal between these two signals. It is desirable to have an adaptive filter for performing the adaptive processing, and to obtain the delay time at the frequency of the overlapping portion corresponding to the non-controlled sound source by performing the Fourier transform processing on the filter coefficients of the adaptive filter. Since the filter coefficient of the adaptive filter reproduces an impulse response detected by the sound collecting means, a delay time corresponding to a frequency overlapping portion can be obtained based on the filter coefficient.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of an equalization system to which a delay time setting method according to the present invention is applied will be described with reference to the drawings.

FIG. 1 is a diagram showing a configuration of an equalization system according to an embodiment to which the present invention is applied. The equalization system shown in FIG. 1 includes a low-pass filter (LPF) 10, a target response setting unit 12, a microphone 14, an adding unit 16, a filter 1
8, speaker 20, 32, LMS algorithm processing unit 2
2. Adaptive filter 24, delay unit 26, switch 28, high-pass filter (HPF) 30, sound source 5 for delay time measurement
0, including a delay time calculation / setting unit 52.

The low-pass filter 10 allows only a low-frequency band, which is a control band, to pass from the input audio signal. The target response setting unit 12 is set with a target response characteristic (impulse response) H, receives an audio signal in the control band output from the low-pass filter 10, and outputs a corresponding target response signal. The microphone 14 is installed at a listening position in the vehicle interior acoustic space, and detects a sound wave that has reached this listening position.

The adder 16 calculates an error between the audio signal output from the microphone 14 and the target response signal output from the target response setting unit 12, and outputs an error signal. The filter 18 convolves the propagation characteristic (transfer characteristic) C of the vehicle interior acoustic system from the speaker 20 to the microphone 14 at the listening position with the audio signal in the control band output from the low-pass filter 10 and uses the reference signal ( (A filtered reference signal). In addition,
In general, a characteristic c ＾ having a characteristic equivalent to the transfer characteristic C of the vehicle interior acoustic system (actually, “＾” is described above “c”, but such a characteristic is used in this specification). Since it cannot be described, it is expressed as “c ＾”), and it is set as a characteristic of the filter 18.

The loudspeaker 20 emits audio sound to the vehicle interior acoustic system based on the audio signal whose characteristics have been corrected by the adaptive filter 24, and a part of the emitted audio sound is collected by the microphone 14. Be sounded.

The LMS algorithm processing unit 22 receives the error signal at the listening position and the filtered reference signal output from the filter 18, and uses these signals to minimize the power of the error signal.
The value of each filter coefficient of the adaptive filter 24 is set using a filtered x (Filtered-x) LMS algorithm. The adaptive filter 24 is a FIR (Finite Impulse Res)
ponse) type digital filter configuration, and LM
Using the filter coefficient vector W (n) set by the S-algorithm processing unit 22, the low-pass filter 10
A predetermined correction process by digital filter processing is performed on the audio signal output from. The corrected audio signal is radiated from the speaker 20 as a control sound source to the acoustic space in the vehicle compartment.

The delay unit 26 delays the input audio signal by a predetermined delay time τ. The delay time τ can be arbitrarily set, and the setting is performed by the delay time calculation / setting unit 52. The switch 28 switches the connection state of the high-pass filter 30 to one of the delay unit 26 side and the delay time measurement sound source 50 side.

The high-pass filter 30 allows only the high-frequency range, which is a non-control band, to pass from the input audio signal. The speaker 32 emits an audio sound corresponding to the non-control band to the vehicle interior acoustic system based on the audio signal whose characteristics have been corrected by the high-pass filter 30, and a part of the emitted audio sound Sound is collected by the microphone 14.

The delay time calculation / setting unit 52 outputs a predetermined signal for delay time measurement from the sound source 50 for delay time measurement, and when a sound wave corresponding to this signal is detected by the microphone 14, the signal is output to the detection signal. The delay time of the delay unit 26 is set based on the delay time. The details of the operation of setting the delay time by the delay time calculation / setting unit 52 will be described later.

The microphone 14 is used as a sound collecting means.
The high-pass filter 30 and the speaker 32 serve as uncontrolled sound sources,
Low-pass filter 10, target response setting unit 12, filter 18, speaker 20, LMS algorithm processing unit 22,
The adaptive filters 24 correspond to the control sound sources, respectively. In addition, the delay unit 26 provides a delay unit with a sound source 50 for delay time measurement.
Corresponds to the measurement signal generation unit, and the delay time calculation / setting unit 52 corresponds to the delay time calculation / setting unit.

The equalizing system of the present embodiment has such a configuration. Next, the operation when the delay time calculation / setting unit 52 sets the delay time of the delay unit 26 will be described.

First, the switch 28 is switched to change the connection state of the high-pass filter 30 from the delay unit 26 to the delay time measuring sound source 50. The switching operation of the switch 28 is performed in response to a switching instruction from the delay time calculation / setting unit 52 or a control unit (not shown).

In the connection state of such a switch,
A predetermined signal for delay time measurement is output from the delay time measurement sound source 50, and a corresponding sound wave is radiated from the speaker 32 to the vehicle interior acoustic space. The delay time calculation / setting unit 52 includes:
A predetermined signal is output from the delay time measurement sound source 50,
When the corresponding detection signal is output from the microphone 14, the frequency characteristic and the group delay time characteristic of the sound pressure level are obtained based on the detection signal.

FIG. 2 is a diagram showing an example of a frequency characteristic and a group delay time characteristic of the sound pressure level obtained by the delay time calculation / setting unit 52. In the present embodiment, the boundary frequency f _{1 at} the boundary position between the control band and the non-control band is
It is set to a predetermined value in advance. Delay time calculation
When the group delay time characteristics as shown in FIG. 2 are obtained, the setting unit 52 obtains the delay time t ₁ at the boundary frequency f ₁ from the obtained characteristics.

Specifically, after a signal is output from the low-pass filter 10, a corresponding sound wave is output from the microphone 14.
, The adaptive filter 24 operates so that the delay time until the delay time reaches the delay time according to the target response characteristic H, and this delay time is eventually given as a known value T ₁ . Therefore, the boundary frequency f _1, the speaker 20
Time T ₁ from the microphone to the microphone 14 and the speaker 3
The delay time t ₁ from the second to the microphone 14 is made equal, that is, the sound wave in the control band (sound wave radiated from the speaker 20) and the sound wave in the non-control band (sound wave radiated from the speaker 32) have almost the same timing In order to reach the microphone 14, the difference (T ₁ −t ₁ ) between these delay times T ₁ and t ₁ may be set as the delay time τ of the delay unit 26.

The delay time calculation / setting unit 52 of the present embodiment
Sets the delay time of the delay unit 26 in this manner, so that near the boundary frequency, the timing at which both the sound wave in the control band and the sound wave in the non-control band reach the listening position is almost the same. Therefore, the sense of incongruity in hearing can be reduced and the sound quality can be improved.

Next, the above-described delay time calculation / setting unit 52
In the following, two specific configuration examples will be described.

FIG. 3 shows an example of the configuration of the delay time calculation / setting unit 52. FIG. 3 shows the configuration of the delay time calculation / setting unit 52 when calculating the delay time by obtaining an impulse response using a time stretched pulse (time stretching pulse). FIG. 3 is a diagram illustrating a configuration. The delay time calculation / setting unit 52 shown in the figure includes an analog-digital (A / D) converter 70, a memory control unit 7
2. Memory 74, averaging unit 76, convolution unit 7
8, an FFT (Fast Fourier Transform) operation unit 80, a delay time calculation unit 82, and a delay time setting unit 84. A time stretched pulse is output from the delay time measuring sound source 50 combined with the delay time calculating / setting unit 52.

The time stretched pulse is a signal whose frequency characteristic H (k) is expressed as follows.

[0034]

(Equation 1)

Here, m is a coefficient indicating the degree to which the phase of each frequency is shifted in the time stretched pulse.
Take any integer value. N is a coefficient that defines the time of generation of the time stretched pulse. Also, k is from 0 to N-
A is an integer up to 1, and a is determined by the third expression included in Expression (1) if m and N are determined. For example, when m = 0, a = 0, so that H (k) = e for all k.
xp (0) = 1, and each frequency component becomes an impulse concentrated without being dispersed.

The actual time-stretched pulse output from the delay time measuring sound source 50 is a signal obtained by performing an inverse Fourier transform on the above-mentioned equation (1), and an example of the signal is shown in FIG. The time-stretched pulse shown in FIG.
= 256, and a signal in which each frequency component is dispersed for a predetermined time corresponding to the value of N and the value of m. Therefore, by setting the value of N to be large and the value of m to be large, the energy of each frequency component can be dispersed over a long period of time. Since the time required for measuring the delay time becomes longer as the occurrence time of the error becomes longer, it is necessary to set appropriate values of N and m within a range where the occurrence time does not become too long.

The A / D converter 70 samples and quantizes the output signal of the microphone 14 at predetermined time intervals, and outputs data of a predetermined number of bits. The memory control unit 72 sequentially stores data output from the A / D converter 70 at predetermined time intervals in the memory 74. When the time stretched pulse is output once, the microphone 14 responds to this time stretched pulse.
Is converted into digital waveform data (hereinafter, this data is referred to as “time-stretched pulse response data”) by the A / D converter 70 and stored in a predetermined area of the memory 74. In the memory 74, such storage areas are secured for L pieces, and when the L time stretched pulses are repeatedly output from the delay time measuring sound source 50, the corresponding time stretched pulse responses are obtained. Data is stored in each of the L storage areas described above.

The averaging section 76 performs averaging processing on the L time-stretched pulse response data stored in the memory 74. The averaged time-stretched pulse response data is q (n), and the i-th response data is q _i
(N)

[0039]

(Equation 2)

## EQU4 ## By averaging L time-stretched pulse response data according to the equation (2), response data from which the influence of sudden noise has been removed can be obtained.

The convolution operation unit 78 calculates the averaged time-stretched pulse response data q calculated by the equation (2).
(N) is convolved with data p (−n) obtained by inverting the time stretched pulse p (n) on the time axis.
FIG. 5 is a diagram showing a signal obtained by inverting the time stretched pulse on the time axis, and shows a waveform obtained by inverting the time stretched pulse corresponding to N = 256 shown in FIG. The data p (-n) actually used in the convolution operation unit 78 is obtained by converting the signal waveform shown in FIG. 5 into digital waveform data, and the sampling interval is A
This is the same as the sampling interval in the / D converter 70.

The convolution operation by the convolution operation unit 78 is performed based on the following equation.

[0043]

(Equation 3)

According to the equation (3), an impulse response is obtained by convolving the time-stretched pulse response signal with a signal obtained by inverting the original time-stretched pulse on the time axis. The convolution operation unit 78 performs a convolution operation using the N pieces of data by sequentially shifting data output from the averaging processing unit 76, and outputs a plurality of operation results.

The FFT operation unit 80 includes a convolution operation unit 78
The calculation result output from the FFT (Fast
Fourier Transform) calculation is performed to obtain the frequency characteristics and group delay time characteristics of the sound pressure level as shown in FIG. The calculation result of the frequency characteristic of the sound pressure level and the calculation result of the group delay time characteristic are output to the delay time calculation unit 82.

The delay time calculating section 82 has an FFT calculating section 80
The delay time to be set in the delay unit 26 is calculated based on the calculation result of the frequency characteristic of the sound pressure level and the calculation result of the group delay time characteristic input from. Specifically, the delay time calculation unit 82 previously stores the above-described boundary frequency f ₁ and the delay time T ₁ from the speaker 20 to the microphone 14 as data, and obtains the boundary frequency f ₁ from the group delay time characteristics. And read the delay time t ₁ at (T ₁ −t
By calculating ₁ ), the delay time τ of the delay unit 26 is obtained. The delay time setting unit 84 sets the delay time of the delay unit 26 based on the delay time τ calculated by the delay time calculation unit 82.

Configuration Example 2 of Delay Time Calculation / Setting Unit 52 FIG. 6 is a diagram showing a configuration of the delay time calculation / setting unit 52 in the case where an impulse response is obtained using an adaptive filter to calculate a delay time. The delay time calculation / setting unit 52 shown in the figure includes an FFT operation unit 80, a delay time calculation unit 82, a delay time setting unit 84, an adaptive filter 90, an LMS (Least Me
an Square) An algorithm processing unit 92 and an adding unit 94 are included. Among them, the FFT operation unit 80, the delay time calculation unit 82, and the delay time setting unit 84 perform basically the same operations as those shown in FIG. 3 described above, and thus detailed description is omitted here. . The delay time measurement sound source 50 combined with the delay time calculation / setting unit 52 outputs white noise (white noise).

The adaptive filter 90 has a FIR (Finite Imp
ulse response) type digital filter, and uses the filter coefficient vector W (n) set by the LMS algorithm processing unit 92 to convert a white noise signal input from the delay time measurement sound source 50. A predetermined adaptive process is performed.

The LMS algorithm processing unit 92
Controls the filter coefficient vector W (n) of the adaptive filter 90 so that the power of the error signal e obtained by subtracting the output signal of the adaptive filter 90 from the output signal of the microphone 14 by the adder 94 is minimized. Therefore,
The output signal of the microphone 14 and the output signal of the adaptive filter 90 are substantially the same, and the filter coefficient vector W (n) of the adaptive filter 90 has substantially the same characteristics as the impulse response detected by the microphone 14.

Therefore, the FFT calculation unit 80 performs the FFT calculation based on the filter coefficient vector W (n) of the adaptive filter 90, thereby obtaining the above-described sound pressure level frequency characteristics and group as shown in FIG. Delay time characteristics are required.

Thereafter, the delay time calculating section 82 calculates the delay time t _{1 at the} boundary frequency f ₁ from the group delay time characteristics.
Is read to calculate (T ₁ −t ₁ ), the delay time τ of the delay unit 26 is obtained, and the delay time of the delay unit 26 is set by the delay time setting unit 84 based on the delay time τ. You.

As described above, the equalizing system according to the present embodiment provides the delay time T ₁ from the loudspeaker 20 to the microphone 14 and the loudspeaker 32 at the boundary frequency f ₁ (frequency at the boundary position between the control band and the non-control band). Since the difference (T ₁ −t ₁ ) from the delay time t ₁ to the microphone 14 is set as the delay time τ of the delay device 26, the sound wave of the control band (the speaker 2) is near the boundary frequency f _1.
The timing at which both the sound wave radiated from 0 and the sound wave in the non-control band (the sound wave radiated from the speaker 32) arrives at the listening position can be made substantially coincident with each other. Can be achieved.

The present invention is not limited to the above-described embodiment, and various modifications can be made within the scope of the present invention. For example, in the above-described embodiment, the sound waves in the control band and the non-control band are output using the two speakers 20 and 32, however, one speaker may be omitted. FIG. 7 is a diagram showing a configuration of a modified example of an equalization system in which one speaker is omitted. The configuration of the equalization system shown in FIG. 7 is different from the configuration shown in FIG. 1 in that a mixing section 34 is provided in front of the speaker 20 so that an audio signal output from the adaptive filter 24 and an audio signal output from the high-pass filter 30 are output. The difference is that the audio signal is mixed with the signal, the mixed audio signal is input to the speaker 20, and the speaker 32 is omitted. With this configuration, the present invention can be applied to an audio system that reproduces the entire reproduction frequency band with one speaker.

In the configuration example 1 of the delay time calculation / setting unit 52 in the above-described embodiment, a time stretched pulse (see FIG. 4) is input to each speaker, and the output signal from the microphone 14 is An impulse response was obtained by convolving a signal obtained by inverting the time-stretched pulse on the time axis (see FIG. 5), but by simply inputting the impulse without using the time-stretched pulse. May be obtained. In this case, the convolution operation unit 78 can be omitted to simplify the configuration.

In the above-described embodiment, a case has been described where the frequency component of the signal in the control band and the frequency component of the signal in the non-control band overlap in some frequency regions. If there is no overlap in the frequency band, an average group delay time of the signal in the non-control band is obtained, and this is set as t ₁ , and the speaker 20,
It is desirable to calculate the difference (T ₁ −t ₁ ) between the delay times of the microphones 32 and the microphone 14 and set the delay time τ of the delay unit 26.

Further, in the above-described embodiment, the case where the delay time setting method of the present invention is applied to an audio system for a vehicle has been described. However, the present invention is not limited to this. It can also be applied to audio systems and the like.

[0057]

As described above, according to the present invention, both the sound wave radiated from the control sound source and the sound wave radiated from the non-control sound source are heard near the frequency at the boundary between the control band and the non-control band. Since the timing of reaching the position can be made substantially coincident, it is possible to reduce a sense of incongruity in hearing and improve sound quality.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration of an equalization system according to an embodiment.

FIG. 2 is a diagram illustrating an example of a frequency characteristic and a group delay time characteristic of a sound pressure level obtained by a delay time calculation / setting unit.

FIG. 3 shows a delay time calculation when a delay time is calculated by obtaining an impulse response using a time stretched pulse.
FIG. 3 is a diagram illustrating a configuration of a setting unit.

FIG. 4 is a diagram illustrating an example of a time stretched pulse.

FIG. 5 is a diagram showing a signal obtained by inverting a time stretched pulse on a time axis.

FIG. 6 is a diagram illustrating a configuration of a delay time calculating / setting unit when calculating an impulse response using an adaptive filter to calculate a delay time.

FIG. 7 is a diagram illustrating a configuration of a modification example of an equalization system in which one speaker is omitted.

FIG. 8 is a diagram showing a basic configuration of an equalization system applied to an in-vehicle audio system.

FIG. 9 is a diagram illustrating a configuration of an equalization system that performs an adaptive equalization process only on bass sounds.

FIG. 10 is a diagram illustrating a conventional method of setting a delay time.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Low-pass filter (LPF) 12 Target response setting part 14 Microphone 16 Adder part 18 Filter 20 and 32 Speaker 22 LMS algorithm processing part 24 Adaptive filter 26 Delayer 28 Switch 30 High-pass filter (HPF) 50 Sound source for delay time measurement 52 Delay time Calculation / setting section

Claims (4)

[Claims] 1. A sound collecting means set at a listening position in an acoustic space; and a sound collecting means installed in the acoustic space, wherein at least a frequency characteristic can be changed and an output sound reaches the sound collecting means. A control sound source whose delay time is known, a non-control sound source which is installed in the acoustic space and whose frequency characteristics cannot be changed, and a delay unit which is connected to a stage preceding the non-control sound source and whose delay time can be set. Measurement signal generation means for generating a predetermined delay time measurement signal radiated from the uncontrolled sound source; and the sound collection means corresponding to the sound wave of the delay time measurement signal radiated from the uncontrolled sound source By analyzing the signal output from the sound source, the overlapped portion of the frequency included in the sound wave radiated from each of the uncontrolled sound source and the control sound source is adjusted so as to substantially coincide with the timing of reaching the sound collecting means. Delay time setting method characterized by comprising a delay time calculation and setting means for setting a delay time of the delay means. 2. The delay time calculating / setting unit according to claim 1, wherein the delay time calculating / setting unit analyzes a signal output from the sound collecting unit in response to a sound wave of the delay time measuring signal emitted from the uncontrolled sound source. Calculating a delay time corresponding to the frequency of the overlapping portion, and setting the delay time of the delay unit such that the calculated delay time matches a known delay time corresponding to the control sound source. Characteristic delay time setting method. 3. The delay time measurement signal generated by the measurement signal generation means according to claim 2, wherein the delay time measurement signal is a time stretched pulse. By performing convolution operation on a signal obtained by inverting the time-stretched pulse on the time axis, and performing a Fourier transform process on the convolution operation result, the delay time at the frequency of the overlapping portion corresponding to the non-controlled sound source A delay time setting method characterized in that: 4. The delay time measurement signal generated by the measurement signal generation means according to claim 2, wherein the delay time measurement signal generated by the measurement signal generation means is a white noise signal. And an adaptive filter that receives the output signal of the delay time measurement signal and the output signal of the sound collection unit and performs adaptive processing so that the power of an error signal between these two signals is minimized. A delay time setting method for performing a Fourier transform process on a filter coefficient of an adaptive filter to obtain a delay time at a frequency of the overlapping portion corresponding to the uncontrolled sound source.