JPH05268700A - Stereo listening aid device - Google Patents

Abstract

PURPOSE:To hear a sound desired to be listened to from mixed noise sound based on a sound direction sense by using an audio signal so as to correct and reproduce the signal while emphasizing presence of the signal. CONSTITUTION:An audio signal 101 is inputted to a sound input section 102, converted into a digital signal 105 by an A/D converter section 104 and inputted to a presence emphasis section 106. The emphasis section 106 applies sound processing to emphasis presence and a 2-channel stereo signal is outputted. The signal 107 is given to a listening sense characteristic correction section 108, in which the correction in matching with a listening impediment is implemented independently of left and right channels. Then a stereo signal 109 outputted from the correction section 108 is converted into an analog signal at a D/A converter 110 and reproduced as an output audio signal 112 by an audio reproduction section 111. The correction section 108 amplifies the waveform with a gain in response to the degree of hearing impediment of left and right ears. Then a discrimination function to hear out a sound desired to be listened to from mixed noise sound is extracted based on a sound direction sense by both-ear listening provided substantially to a human being.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a hearing aid, and more particularly to a digital hearing aid that employs multiple channel inputs for audio signal processing.

[0002]

2. Description of the Related Art Conventional hearing aids such as hearing aids are mainly for one channel for one ear, and most of them are for amplifying and reproducing input voice by an analog amplifier. Digital hearing aids, which have become popular in recent years, are capable of realizing functions such as AGC, suppression of instantaneous large-amplitude input, gain control for each band, and selection of characteristics that match environmental conditions, using digital signal processing technology. Became. In addition, the two-input method uses a stereo type that amplifies the sound input from the left and right microphones and reproduces from both ears, and one input as the sound input and the other input as the noise reference input as a noise countermeasure. There is also one using a process of removing the estimated noise component from the input signal.

[0003]

One of the major problems of the current hearing aids represented by hearing aids is noise. Current hearing aids amplify not only the sound you want to hear, but also the noise you do not want to hear. Therefore, conversation in a noisy environment is very difficult for a hearing aid user to hear. As a countermeasure against noise in a hearing aid, there is an approach of removing noise components by digital signal processing. As an example,
There is a method in which a noise reference microphone is provided in addition to the voice microphone, and the estimated noise component is subtracted from the main input signal.

Most of the conventionally used hearing aids are monophonic ones with one input and one output. However, with a monaural hearing aid, only one ear can be heard, so by using both ears, it is possible to bring out the azimuth of sound that should be inherent in humans and the ability to discriminate the desired sound from noise. I could not do it. With stereo type hearing aids that have been used in recent years, it is possible to listen to both ears by compensating for the damage in both ears independently, and it is considerably easier to hear than the above-mentioned mono type hearing aid. However, currently used stereo type hearing aids have a structure with only two inputs and outputs, and do not have a process that positively utilizes the effect of both ears that humans originally have. For this reason, even the current stereo hearing aids have not been sufficiently effective for binaural listening.

An object of the present invention is to perform signal processing for enhancing the sense of presence of a sound signal input from a plurality of channels while independently correcting the auditory characteristics of the left and right ears of the user. The purpose of this is to provide a device that actively draws out the ability that humans should have, such as the sense of direction of sound by binaural listening and the function of discriminating the desired sound from the noise.

[0006]

In order to achieve the above object, a means for inputting audio signals of a plurality of channels, a means for emphasizing the realism of the signal using the input audio signals,
Means for independently correcting the characteristics of the stereo signal with the enhanced sense of presence are provided in the order of construction.

[0007]

Various modifications are conceivable in the present invention, and the operation of typical means will be described.

A voice input unit for inputting voice signals of at least a plurality of channels, a realism emphasizing unit for emphasizing the realism of the signal by using the inputted voice signals, and a stereo signal outputted from the realism emphasizing unit. By configuring the auditory characteristic correction unit that corrects the characteristics independently on the left and right and the audio reproduction unit that reproduces the stereo signal processed by the auditory characteristic correction unit in the above order, audio is input to the multi-channel audio input unit. When input, the input signals of each channel are combined into a stereo signal with the presence enhanced in the presence enhancing section, and in the next auditory characteristic correction section, a stereo signal with the presence of the signal is enhanced in the presence enhancing section. On the other hand, it is possible to correct the auditory characteristics of the user independently.

Therefore, according to the present invention, by performing signal processing for emphasizing the realism of the signal to the audio signals input from a plurality of channels, the direction of the sound due to the binaural listening that should be originally possessed by human beings. It is possible to positively bring out the ability such as a discrimination function of distinguishing a desired sound from feelings and noises, and it is possible to independently correct the left and right ears of the user in response to different hearing impairments.

[0010]

[Embodiment] To simplify the explanation, the number of input channels is set to 2 in the following, but even in the case of 3 or more channels, each master tape recorded in multi-channels is mixed down for 2-channel audio. The channels can be combined in the same way. Also, in order to simplify the explanation, explanation using digital signal processing is given.
It is also possible to perform some or all of them by analog signal or hybrid signal processing.

FIG. 1 is a block diagram for explaining an embodiment of the present invention. In FIG. 1, the voice 101 is input by the voice input unit 102. The input signal 103 input to the voice input unit 102 is converted into a digital signal 105 by the A / D conversion unit 104, and enters the realism emphasizing unit 106. The presence enhancing unit 106 performs processing such that the presence is enhanced by surround processing, and outputs a 2-channel stereo signal 107. The auditory characteristic correction unit 108 independently corrects the stereo signal 107 in which the sense of presence is emphasized in the left and right channels, for example, according to the hearing impairment of the user. The stereo signal 109 output from the hearing characteristic correction unit 108 is converted into an analog signal by the D / A conversion unit 110, and is reproduced as the output sound 112 by the sound reproduction unit 111.

Each part will be described in detail below. The voice input unit 102 generally uses a plurality of microphones. The microphones are preferably placed near both ears of the user in order to maximize the effect of hearing with both ears, but it is also possible to place them in other places. Further, the number of microphones does not have to be two as long as input of a plurality of channels can be obtained. Further, in addition to using a microphone as a sound input means, for example, by using a terminal that can be connected to an audio device such as a television, it is possible to input an audio signal output from these audio devices. By providing such a means, it is possible to listen to audio such as stereo broadcasts of televisions, etc. with a high sense of realism.

FIG. 2 is an explanatory diagram of an example of realization as an example of the realism emphasizing section 106 when the input signal is two channels. In the present embodiment, the enhancement of the realism is realized by the surround processing. Regarding the surround processing, the paper “Technical file STC-020 of the Japan Electronic Machinery Manufacturers Association” and Niimi et al. “On sound field processing for audio / video reproduction”
It is described in detail in Proceedings of the Acoustical Society of Japan (Sho 61-10).

In FIG. 2, 201 and 201 'are input signals of two channels, 202 and 202' are amplitude adjusting sections, and 20.
4, 204 'are signal delay units. The input signal 201 input from one of the channels is subjected to signal amplitude correction by the amplitude adjustment unit 202 and then enters the signal delay unit 204. The signal delay unit 204 corrects the phase of the signal by delaying the input signal 203. The signal 205 whose amplitude and phase have been adjusted is the input signal 201 'of another channel.
Is synthesized with. Here, the correction amounts in the amplitude adjusting units 202 and 202 'and the signal delay units 204 and 204' are emphasized to enhance the sense of presence so that the effect of listening to both ears of the user can be sufficiently obtained.
It can be adjusted freely. This adjustment can be realized by, for example, a method of inputting a reference signal from the audio input unit 102 and adjusting so that the user listening to the reproduced signal can obtain an optimum realistic sensation. Further, it is also possible to preset the adjustment amount in a plurality of use environments and select it according to the use environment. In this embodiment, the presence enhancement process is realized by correcting both the amplitude and the phase, but the presence enhancement process may be performed by correcting only the amplitude or the phase.

It is also possible to realize the processing of the realism emphasizing section 106 in the frequency domain. FIG. 3 shows the presence enhancing unit 10.
It is an embodiment in which the processing of No. 6 is realized in the frequency domain. Figure 3
, 302 and 302 'are waveform cutouts, and 30
4, 304 'are Fourier transform units, 306, 306' are buffer memory units, and 308, 308 'are amplitude adjusting units. The input signal 301 input from one channel is input to the waveform cutout unit 302. Waveform cutting section 30
Reference numeral 2 is for cutting out a waveform section (frame) for analyzing spectrum information from an input signal, and cutting out a section of about several tens ms at regular intervals. The cut-out section signal waveform 303 is converted into spectrum data by the Fourier transform unit 304.

Here, after the cut-out waveform is multiplied by a window function such as a Hanning window, which is normally used, zero data is embedded before and after it to obtain data of factorial factor of 2, thereby obtaining a fast Fourier transform (FFT). ) Can be used, and high-speed data processing is realized. In addition, as in this embodiment, 2
When performing Fourier transform on a channel signal, the processing amount can be reduced by using a method of simultaneously obtaining the Fourier coefficient of two signals by utilizing the redundancy of Fourier transform. This two-signal simultaneous FFT method is a method proposed by Nakano, and is described in detail in Nakano et al., "Analysis of Ship Running Sound by Fast Fourier Transform," Proceedings of Acoustical Society of Japan (Sho 43-11). The outline is given here.

The FFT normally performs a complex number operation,
Since the actual signal is a real number, redundancy occurs. When two types of input data are X _i and Y _i (i = 0 to N−1) and their Fourier transforms are A _k and B _k (k = 0 to N−1), two input data About X _i and Y _i

[0018]

## EQU1 ## When Z _i = X _i + jY _{i is set} and the Fourier transform C _k of Z _i is _obtained, as is clear from the definition of the Fourier transform,

[0019]

## EQU2 ## C _k = A _k + jB _k . On the other hand, if * represents a conjugate complex number,

[0020]

Since there is a relationship of A _Nk = A _k * B _Nk = B _k *,

[0021]

(4) C_Nk= A_Nk+ JB_Nk  = A_k* ＋ jB_k* Therefore,

[0022]

## _EQU5 ## C _k + C _Nk = (A _k + A _k *) + j (B _k + B _k *) = 2Re (A _k ) + 2jRe (B _k ).

[0023]

[6] _{C k -C Nk = (A k} + A k *) + j (B k -B k *) = -2Im (B k) + 2jIm (A k) is obtained. However, Re represents a real part and Im represents an imaginary part. That is, if C _k is obtained, then A _k and B _k can be easily obtained. The calculation amount required to obtain C _k is A _k or B _k
It is exactly the same as the amount of calculation for obtaining, and because the amount of calculation for the C _k get A _k, B _k is negligible compared to the amount of calculation for obtaining the C _k, conventional Two kinds of signals can be simultaneously Fourier-transformed with an almost same amount of calculation.

The spectrum data 305 converted by the Fourier transform unit 304 is stored in the buffer memory unit 306. The buffer memory unit 306 is configured to be able to store the spectrum data of the past several frames, and outputs the spectrum data of the past several frames with respect to the input spectrum data to realize the signal delay processing. The spectrum data 307 output from the buffer memory unit 306 is amplitude-corrected by the amplitude adjusting unit 308 and then combined with the spectrum data 305 'of another channel. Although the directivity of the sound wave has a characteristic that depends on the frequency, since the amplitude adjusting unit 308 is a process in the frequency domain, it is also possible to correct the amplitude that depends on the frequency.

The output of the realism emphasizing section 106 is then sent to the auditory characteristic correction section 108, where it is subjected to characteristic correction and then converted into a waveform region by inverse Fourier transform. When the correction by the auditory characteristic correction unit 108 is performed in the waveform region, the inverse Fourier transform may be performed before that.

The hearing characteristic correction unit 108 will be described.
The simplest method for correcting hearing impairment is to amplify the waveform independently of the left and right with a gain according to the degree of hearing impairment of the left and right ears. However, this method cannot correct even the frequency characteristic of the obstacle.

FIG. 4 is an explanatory diagram of an embodiment in which the hearing characteristic correction unit 108 is realized in consideration of the frequency characteristic of hearing impairment. The auditory characteristic correction unit processes left and right channels, but since both channels have exactly the same configuration, only one channel is shown in this figure. It should be noted that the description of the present embodiment will be made assuming that the realism emphasizing unit 106 is processed in the waveform region as in the method described in FIG.

In FIG. 4, reference numeral 402 denotes a waveform cutout portion,
404 is a Fourier transform unit, 406 is a frequency characteristic correction unit,
Reference numeral 408 is an inverse Fourier transform unit, and 410 is a waveform synthesis unit. The signal 401 output from the realism emphasizing unit 106 is input to the waveform cutout unit 402. Waveform cutting section 40
2 is to cut out a waveform section for analyzing spectrum information from the input signal, and cut out a section of about several tens ms at a constant interval. The cut-out signal waveform 403 is converted into spectrum data by the Fourier transform unit 404. As described above, the Fourier transform unit 404 can significantly reduce the processing amount by adopting the fast Fourier transform (FFT) or the two-signal simultaneous FFT. Further, as in the embodiment shown in FIG. 3, when the realism emphasizing unit 106 has already converted the spectrum data, the waveform cutting unit 402 and the Fourier transform unit 404 are unnecessary.
The spectrum data 405 converted into the frequency domain enters the frequency characteristic correction unit 406. Frequency characteristic correction unit 406
Performs a filtering process in the frequency domain to correct the deterioration of the frequency characteristic due to the hearing loss of the user. The correction of the frequency characteristic can be easily applied individually by inputting the reference signal for each frequency at the same level on the left and right and adjusting the filter parameters so that the user can hear at the same level on the left and right. Spectral data 407 with corrected frequency characteristics
Is transformed into the time domain again by the inverse Fourier transform unit 408, and the waveform synthesizing unit 410 synthesizes the waveform 409 cut out at the analysis frame period as the speech waveform 411.

Further, the inverse Fourier transform unit 408 can also greatly reduce the amount of processing by adopting the inverse transform using the two-channel simultaneous FFT method. That is, two types of Fourier-transformed signals are converted into A ′ _k , B ′ _k (k = 0 to N
−1), and the respective inverse Fourier transforms are X ′ _i and Y ′ _i.
Then,

[0030]

Equation 7] _{C 'k = A' k +} jB 'k Distant, C' when seeking inverse Fourier transform Z _'i of _k,

[0031]

The [number 8] _{Z 'i = X' i +} jY 'i. Here, since X ′ _i and Y ′ _i are obviously real numbers, two signals can be separated by taking the real part or imaginary part of Z ′ _i . Since the amount of calculation required to obtain Z ′ _i is exactly the same as the amount of calculation required to obtain X ′ _i or Y ′ _i , it is necessary to simultaneously perform the inverse Fourier transform on two types of signals with an amount of calculation that is almost the same as the conventional one. You can

The effect of this embodiment is that, by the surround processing in the realism emphasizing section 106, the sense of direction of the sound due to the binaural listening which should be originally possessed by the human being and the discrimination function of distinguishing the desired sound from the noise are provided. The ability can be positively drawn out, and the hearing characteristic correction unit 108 can correct the hearing impairment of the user independently for the signal in which the sense of presence is emphasized.

[0033]

As described above, according to the present invention, not only the left and right characteristics can be corrected independently by the correction processing of the auditory characteristics, but also the realism of the signal is emphasized with respect to the audio signals input from a plurality of channels. By applying the signal processing that should
It is possible to proactively bring out the ability that humans should have, such as the sense of direction of sound by binaural listening and the function of discriminating the desired sound from the noise.

[Brief description of drawings]

FIG. 1 is a block diagram of a system configuration according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram of a realism emphasizing unit.

FIG. 3 is an explanatory diagram of a realism emphasizing unit.

FIG. 4 is an explanatory diagram of a hearing characteristic correction unit.

[Explanation of symbols]

101 ... Input voice, 102 ... Voice input unit, 104 ... A /
D conversion unit, 106 ... Reality enhancement unit, 108 ... Hearing characteristic correction unit, 110 ... D / A conversion unit, 112 ... Sound reproduction unit.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yoshihito Nahito 1-280, Higashi Koigokubo, Kokubunji, Tokyo Inside the Central Research Laboratory, Hitachi, Ltd.

Claims (16)

[Claims] 1. A voice input unit for inputting voice signals of a plurality of channels, a presence enhancing unit for enhancing the presence of signals using the input voice signals, and a stereo output from the presence enhancing unit. It is characterized in that it includes an auditory characteristic correction unit that corrects the characteristics of the signal independently left and right, and an audio reproduction unit that reproduces the stereo signal processed by the auditory characteristic correction unit, each of which is configured in the above-described configuration order. Hearing aids. 2. The hearing aid device according to claim 1, wherein the hearing characteristic correction unit corrects the frequency characteristic of the hearing impairment of the user. 3. The hearing aid device according to claim 1, wherein the hearing characteristic correction unit realizes correction of the frequency characteristic using a digital filter. 4. The Fourier transform unit for performing a Fourier transform of a signal before the auditory characteristic correction unit according to claim 1 or 2, wherein the auditory characteristic correction unit converts the signal subjected to frequency analysis by the Fourier transform unit. A hearing aid that compensates for this. 5. The hearing aid device according to claim 1, wherein the auditory characteristic correction unit corrects the signal in a waveform region. 6. The hearing aid device according to claim 1, wherein the correction in the waveform region in the hearing characteristic correction unit is realized by adjusting the gains of the left and right channels and changing the signal level. 7. The realism emphasizing unit according to claim 1, wherein the realism emphasizing unit has a Fourier transforming unit for converting the input audio signal into a frequency domain for each frame. And a hearing aid device that performs the process of the hearing characteristic correction unit on the signal converted into the frequency domain. 8. The hearing aid device according to claim 1, 2, 3, 4, 5, 6 or 7, further comprising means for giving a delay time to an input signal of each channel in the presence enhancing section. 9. A hearing aid device according to claim 1, 2, 3, 4, 5, 6 or 7, further comprising means for adjusting the amplitude of the input signal of each channel in the presence enhancing section. 10. The method of claim 1, 2, 3, 4, 5, 6 or 7.
2. The hearing aid device according to claim 1, wherein the realism emphasizing section has means for adjusting the amplitude of the input signal of each channel and means for giving a delay time. 11. The realism emphasizing section according to claim 7, further comprising a buffer memory section for storing a predetermined number of frames for each frame of the result of frequency analysis by the Fourier transform section, and storing the result in the buffer memory section. Hearing aid having means for implementing the delay of the signal by using a past analysis frame that has been created. 12. The Fourier transform unit as set forth in claim 4, 7 or 11, wherein voices input from a plurality of channels are collected for every two channels, and signals of each two channels are real part in one and imaginary part in the other. A hearing aid having means for performing a Fourier transform, which is regarded as a complex number signal sequence, and means for separating the Fourier transform of two signals from the result of the Fourier transform. 13. The hearing aid device according to claim 1, further comprising means for varying the amplitude adjustment amount according to the frequency band of the input signal of each channel in the realism emphasizing unit. 14. The hearing aid device according to claim 1, wherein a microphone is used as the input means. 15. The hearing aid device according to claim 1, wherein the microphone is installed near both ears of the user. 16. The hearing aid device according to claim 1, wherein the input means has a terminal connectable to an output terminal of an audio device.