JPS6320240Y2 - - Google Patents

Description

[Detailed description of the invention] This invention relates to the improvement of hearing aid devices, and in particular, the invention automatically changes the frequency characteristics depending on ambient noise.
To provide a hearing aid device that controls a hearing aid so as to reduce "noisiness" caused by noise and reliably discriminates between noise and conversational speech.

Traditionally, when hearing aid wearers judge that ambient noise is extremely noisy, they
In order to avoid this, people manually adjust the volume adjuster and frequency response adjuster included in their hearing aids. It would be more convenient to perform this operation automatically, but in order to do this, the hearing aid must first have the ability to judge the nature of the input sound and distinguish whether it is noise or conversational sound. is necessary.

Conventionally, this type of identification means (a) when an input sound of a higher level than expected is received; or (b) when low frequency and/or high frequency spectral components, which are not often included in conversational speech, reach a higher level than expected in advance.

A method of identification using .

However, in general, hearing aids must use the wearer's own voice as input sound as well as the voice of the conversation partner. Therefore, apart from situations where you hold out your hearing aid right in front of the person you are talking to, for example, you should not wear your hearing aid in your pocket, in your ear, or in your ear canal. During a conversation in this state, the wearer's own voice is likely to be louder, depending on the distance to the other party, and the difference in level between the user's own voice and that of the other's voice is expected to be several tens of decibels. Furthermore, if we consider the difference of several tens of dB between loud voices and soft voices, the expected high level in (a) above should be set to a relatively high value, taking into account the level difference mentioned above. become.

Next, regarding the above (b) identification method that uses components that are not often included in conversational speech, since even conversational speech does not contain such components at all, (b) As with other cases, it is necessary to consider the distance between you and the other party and whether you speak loudly or quietly.

The above consideration necessarily narrows the range of levels that can be distinguished as noise or conversational speech,
In addition, we will be forced to limit the distance between us and our conversation partners.

This invention solves the above-mentioned conventional problems, and is characterized by being able to more reliably distinguish between noise and conversational sounds by focusing on their characteristics over time. In other words, although there are special cases such as impact sounds and singing voices, this characteristic over time is similar to the street noise on the main street shown in Figure 1A, the noise inside the subway train shown in Figure 1B, or the noise caused by passing airplanes or trains. In contrast, in conversational speech, the level is continuous and has little variation.
As shown in Figure 2, it is extremely intermittent. FIG. 2 shows, as an example, the manner in which the level changes in a conversational voice asking, ``Which is faster, Dempo or Tegami?''.

This invention was made by focusing on whether the level of the input sound is continuous or intermittent to distinguish between noise and conversational speech. Furthermore, in addition to this level change, it is also conceivable to perform identification by taking into consideration the magnitude of the level.

For this reason, in the present invention, a microphone, a detection circuit connected to the output of the microphone, and a detection circuit connected to the output of the microphone in parallel with the detection circuit, are detected by the detection signal obtained from the detection circuit. a hearing aid amplification circuit in which low frequency characteristics and/or high frequency characteristics are controlled; the detection circuit includes a nonlinear amplifier that amplifies the electrical signal from the microphone; and rectifies the output of the nonlinear amplifier; It is composed of a rectifier circuit for leveling and a time constant circuit connected to the output of the rectifier circuit, and the time constant circuit is set to have a long rise time and a short release time.

An embodiment of this invention will be described below based on the drawings.

FIG. 3 shows a detection circuit for identification, which embodies the function of detecting not the average height of the input sound level but the degree of change in level. That is, in order to focus on the intermittency of the electrical signal from the microphone 17 rather than its average height, it is amplified by a nonlinear amplifier 1 having characteristics as shown by curve b in FIG. After rectifying it, it is added to the time constant circuit 3. This time constant circuit 3 is
It has the characteristics of a long rise time and a short release time. The detection (identification) signal thus obtained becomes a signal weighted according to the degree of continuity or intermittentness of the original input sound.

If the rise time of the time constant circuit is short and the release time is long, this detection signal is a signal close to the maximum value of the signal from the nonlinear amplifier 1 (FIG. 6A), corresponding to the audio signal in FIG. When the rise and release times are equal, the average value signal (6th B
(Fig. 6C), and by making the rise time longer than the release time, a signal representing the degree of continuity or intermittentness (Fig. 6C) can be obtained.

FIG. 5 shows another modification of the detection circuit, which is composed of a variable amplifier 4 (non-linear circuit), a rectifier circuit 5, and a time constant circuit 6. The output of the time constant circuit 6 is used as a feedback signal to the variable amplifier 4. It also controls the amplification degree of the variable amplifier 4 and realizes non-linear operation. The time constant circuit 6 is set to have a long rise time (several seconds to tens of seconds) and a short release time (several seconds). As in the case of FIG. 3, the detection signal from the time constant circuit 6 is also a signal weighted according to the degree of continuity or intermittence of the original input sound.

In addition, considering the characteristics such as curves a, b, and c in FIG. 4 showing the input-output relationship of the nonlinear amplifiers 1 and 4 in FIGS. 3 and 5, it can be seen that according to the present invention, the input signal It can be seen that in addition to physical characteristics, level factors can also be used as discrimination factors. That is, if the characteristic of curve a, which has a point (level) at which non-linear amplification occurs is different from that of curve b, is adopted, only the degree of continuity or intermittentness of the input sound is detected, regardless of the level of the input sound. This is intended to allow listening with the minimum necessary frequency bandwidth to maintain the intelligibility of conversation at all times, even if the input sound level is not very high.

In the case of curve c, whether the input sound is continuous or intermittent is detected only for input sounds of a certain level or higher. This means that if the input sound is below a certain level, it will not be too noisy, so whether the input sound is noise or conversational sound, you will listen to it. Therefore, it is preferable to configure the nonlinear amplifiers 1 and 4 so that the point at which nonlinear amplification occurs in the input/output relationship is variable, so that the degree of the above-mentioned effect can be adjusted.

FIG. 7 shows an embodiment of a hearing aid device including the detection circuit 11 as described above, and 12 is an amplifier circuit as the hearing aid. This hearing aid amplifier circuit 12 is
It has the functions of a normal hearing aid, but it also converts audio signals into low frequencies, such as below 600Hz, mid-range,
For example, a low frequency filter 13A, a middle frequency filter 13B, and a high frequency filter 13C are provided for dividing the frequency range from 600 to 3000 Hz, for example, 3000 Hz or higher, and amplifiers 14A, 14B, and 14C are connected to the filters, respectively. . 15 is a preamplifier, and 16 is a main amplifier. The output of the microphone 17 is connected to the input side of the detection circuit 11 and the preamplifier 15, and the output of the main amplifier 16 is configured to be applied to the earphone 18.

The output of the detection circuit 11 is applied to low-frequency and high-frequency amplifiers 14A and 14C to control their amplification degrees.

With the above configuration, when a detection signal indicating noise is applied from the detection circuit 11, the amplifier 1
The amplification of 4A and 14C is lowered, changing the frequency characteristics of the hearing aid as a whole, leaving at least the frequency breadth necessary for conversational speech, and reducing "noisyness." Furthermore, when the overall level of the input sound is low or it is identified as conversational sound, a wide range of audible sounds ranging from low to high frequencies can be obtained.

Note that, as described above, the nonlinear amplifier 1 in the detection circuit 11 can also make the point at which nonlinear amplification occurs in the input/output relationship variable.

As mentioned above, this invention improves the performance of reliably distinguishing between noise and conversational voices and automatically adjusting the frequency characteristics of hearing aids, and has great practical effects.

[Brief explanation of the drawing]

Figures 1A and 1B are diagrams illustrating changes in the level of noise, Figure 2 is a diagram illustrating changes in the level of conversational speech, and Figure 3 is an implementation of a detection circuit that is part of this invention. 4 is a characteristic diagram, FIG. 5 is a block diagram of another embodiment, and FIGS. 6A, 6B, and 6C are diagrams for explaining the detection circuit output, respectively. FIG. 7 is a block diagram of an embodiment of this invention. 1... Nonlinear amplifier, 2... Rectifier circuit, 3...
Time constant circuit, 11...detection circuit, 12...hearing aid amplification circuit, 17...microphone.

Claims (1)

[Claims for Utility Model Registration] (1) A microphone 17, a detection circuit 11 connected to the output of the microphone 17, and a detection circuit 11 connected to the output of the microphone 17 in parallel with the detection circuit 11, The hearing aid amplifier circuit 12 has a hearing aid amplifier circuit 12 whose low frequency characteristics and/or high frequency characteristics are controlled by the obtained detection signal. It is composed of a linear amplifier 1, a rectifier circuit 2 that rectifies and levels the output of the non-linear amplifier 1, and a time constant circuit 3 connected to the output of the rectifier circuit 2, and the time constant circuit 3 includes: A hearing aid device characterized by having a long rise time and a short release time. (2) The hearing aid device according to claim 1, wherein the nonlinear amplifier 1 is a variable amplifier controlled by a feedback signal from a time constant circuit 3. (3) The hearing aid device according to claim 1, wherein the point at which nonlinear amplification occurs in the input/output relationship of the nonlinear amplifier 1 is made variable.