JPH0495500A - Hearing aid - Google Patents

Abstract

PURPOSE:To suppress the most offensive sound to a hearing-aid user and to relieve the pain to the user by discriminating between the contact noise of a hard material and a conversation voice by utilizing the difference between their rising waveforms and suppressing only the contact noise. CONSTITUTION:A peak detecting circuit 5 generates a pulse the moment an envelope signal 22 has a peak value. The output of a binarizing circuit 4 is inputted to a delay circuit 6 and a timer circuit 7. The circuit 6 delays the input signal 23 by a constant time and outputs the delayed signal. The timer circuit 7 detects the rising time the signal 23 of a circuit 4 and generates a constant-level signal. The output 25 of the circuit 7 and the output 24 of the circuit 5 are inputted to an AND circuit 8 and ANDed. The output 28 of a signal selecting circuit 9 and the output 27 of the circuit 6 are inputted to an AND circuit 10. The output 29 of the circuit 10 is used to control a switch 11 which connects a delay circuit 2 and an earphone 12 and the circuit is opened when the output of the circuit 10 is not zero and closed when zero. The output of the delay circuit 2 is also delayed, so the switch 11 is opened to erase only the contact noise from the input to the earphone 12.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention This invention relates to a hearing aid configured to eliminate sounds that are harsh to hearing aid users, such as metal contact sounds.

"Prior Art" Conventional sound processing methods for eliminating harsh sounds in hearing aids include methods for suppressing excessive sounds and methods for suppressing only high-frequency sounds using a frequency filter.

``Problems to be Solved by the Invention'' However, conventional processing methods have not been able to effectively suppress the sound of contact between hard materials such as metal, which is the most annoying sound for hearing aid users. This is because the contact sound of hard materials includes low frequency components, and the sound pressure is not necessarily high.

An object of the present invention is to provide a hearing aid that includes means for effectively suppressing the contact sound of hard materials, which has not been suppressed by conventional hearing aids.

[Means for Solving Problem R J As a result of measuring a large number of hard material contact sounds and conversation sounds, it was discovered that the rise time of contact sounds is significantly shorter than that of conversation sounds. This invention is based on this discovery,
It distinguishes contact sounds from conversation sounds and suppresses only contact sounds. A hearing aid according to the present invention includes a means for collecting sound, a means for extracting an envelope of the collected sound, and a means for determining the magnitude relationship between the rise time of the extracted envelope and a predetermined reference time. a means for transmitting the collected sound with a delay of more than a reference time; a means for suppressing a part of the delayed sound when it is determined to be small according to the determination result of the magnitude relationship; and means for reproducing the suppressed sound.

"Function" In the hearing aid of the present invention, after collecting sounds and extracting the envelope line of the collected sound, the rise time of the envelope line is compared with a predetermined reference time, and the collected sound is extracted. It is determined whether the sound matches a contact sound or a conversation sound. That is, if the rise time of the envelope line is shorter than the reference time, it is determined that the sound is a contact sound of a hard substance, and if it is longer than the reference time, it is determined that the sound is a conversation sound. In parallel, the collected sounds are transmitted with a certain time delay.

Since the above determination requires a processing time longer than the reference time, the delay time is set to be longer than the reference time. If the determination result is small, only the contact sound is eliminated by suppressing the corresponding portion of the delayed sound. As a result, it becomes possible to suppress contact sounds, which was not possible with conventional hearing aids, eliminate harsh sounds, and alleviate the pain of hearing aid users.

"Embodiment" FIG. 1 shows an embodiment of the present invention, and FIG.
An example of signals of each part in the illustrated embodiment is shown. External sounds are collected by microphone 1 and converted into electrical signals.

This electrical signal has a waveform as shown at 21 in FIG. 2, for example. In the same figure, part 41 is a part where there is no sound and the signal level is zero, part 42 is a part of conversation sound, part 43 is a part with no sound, part 44 is a part of the sound of contact with a hard material, and part 45 is a part with no sound. This is the missing part.

A signal 21 output from the microphone 1 enters a delay circuit 2 and an envelope detection circuit 3. The delay circuit 2 delays the signal input from the microphone 1 by a certain period of time and outputs the delayed signal. The delay time is set to a certain value in advance. This is hereinafter referred to as the reference time. This reference time is an intermediate value between the rise time of the hard material contact sound and the rise time of the conversation sound. A specific value is set to 31S, for example.

The output of the delay circuit 2 has a waveform like 30 in FIG.

The envelope detection circuit 3 extracts the envelope of the sound.

As a method for actually realizing the envelope detection circuit 3, for example, the following method may be used. First, the input signal 21 from the microphone 1 is A/D converted, and then a sample value at a certain time is compared with sample values before and after that time. If the central sample value is larger than the other sample values, or if the three values are equal, the central sample value is output. If the center sample value is not large, the previous output value is output. If this is repeated one after another in chronological order, a waveform such as shown in waveform 22 in FIG. 2, which connects only the maximum points of the original waveform, can be obtained. Note that the above processing can be easily realized using existing digital signal processing technology.

The envelope signal 22 enters the binarization circuit 4 and the peak detection circuit. The binarization circuit 4 determines whether the voltage of the envelope signal 22 is below a certain level and is greater than that level.
If it is below the level, it outputs zero, and if it is greater than the level, it outputs a certain non-zero signal (for example, TTL level). The level serving as the criterion is set to a value sufficiently smaller than the average voltage of the envelope signal 22. This may be zero. The output signal from the binarization circuit 4 has a waveform 23 in FIG.

The peak detection circuit 5 generates a pulse of a constant level (same magnitude as the non-zero output of the binarization circuit 84) at the moment the envelope signal 22 takes a peak value. This signal becomes waveform 24 in FIG.

The output of the binarization circuit 4 is the delay circuit 6 and the timer times! ! !
7 is input. In the delay circuit 6, the input signal 2
3 is output with a certain time delay. This delay time is the same as the reference time mentioned above, and its output waveform is waveform 27 in FIG.

Timer times! 37 detects the rising edge of the output signal 23 of the binarization circuit 4, and generates a signal at a constant level (same magnitude as the non-zero output of the binarization circuit 4) for a certain period of time from this moment. This time is also the same as the reference time mentioned above, and the output waveform becomes waveform 25 in FIG.

Output 25 of timer circuit 7 and output 2 of peak detection circuit 5
4 is input to the AND circuit 8, and a logical product is taken. In the conversational sound part, the peak occurs after the reference time after the sound is ordered, so no pulse appears at the output of the AND circuit 8. In the contact sound part, the peak occurs within the reference time even though the sound is generated. As a result, a pulse appears at the output of the AND circuit 8.

As a result, the output of the AND circuit 8 has a waveform as shown in 26 (FIG. 2) in which only pulses corresponding to the peaks of the contact sound appear.

The output 26 of the AND circuit 8 and the output 27 of the delay circuit 6 are
The signal is input to the signal selection circuit 9. The signal selection circuit 9 outputs a zero level signal to the input from the delay circuit 6 after detecting the fall of the signal, and outputs a zero level signal to the input from the AND circuit 8 after detecting the rise of the signal. It outputs a constant zero signal (same size as the non-zero output of the binarization circuit 4), and for example, a flip-flop is set at the rise of the output of the AND circuit 8, and the flip-flop is delayed. It is sufficient to reset it at the fall of the output of the circuit 6. As shown in waveform 28 in FIG. 2, the output of this signal selection circuit 9 is non-zero only during the time when the end of the contact sound is delayed by the reference time from the pulse generation time corresponding to the peak of the contact sound. , and the others are zero.

The output 28 of the signal selection circuit 9 and the output 27 of the delay circuit 6 are input to the AND circuit 10. AND circuit】O II
F- is the same as AND circuit 8. As shown by waveform 29 in FIG. 2, the output of the AND circuit 10 becomes non-zero only during the time when the generation time of the contact sound is delayed by the reference time.

The output 29 of the AND circuit IO is the delay circuit 2 and the earphone 1.
The switch 11 that connects the 2 and 2 is controlled.

The switch 11 is configured such that the circuit opens when the output 29 of the AND circuit 10 is non-zero, and closes when the output 29 is zero.

Since the output from the delay circuit 2 is also delayed by the reference time as shown in waveform 30 in the second round, the above operation opens the switch 11 only while the contact sound is occurring, and only the contact sound is generated. will be deleted from the input of the earphone 12. The input waveform of the earphone 12 is a waveform 31 in FIG.

In the above embodiment, the delay time of the delay circuit 2 is made to match the reference time, but the delay time may have any value as long as it is longer than the reference time. However, in this case, the AND circuit 10 and the switch 11 to synchronize the non-zero portion of the waveform 29 in FIG. 2 with the contact sound portion from the delay circuit 2. The delay time of this new delay circuit may be set as the difference between the delay time of the delay circuit 2 and the reference time.

In the embodiment of FIG. 1, the microphone 1 is a means for collecting sound, the envelope line detection circuit 3 is a means for extracting the sound envelope,
The value conversion circuit 4, the peak detection circuit 5, the timer circuit 7, and the AND circuit 8 are means for comparing the rise time of the envelope with a preset time, and the delay circuit 2 is a means for delaying the sound. 6, signal selection circuit 9 and AND circuit 1o
and the switch 11 constitute a means for suppressing the delayed sound of a portion determined to be low according to the magnitude relationship, and a means for the earphone 12 to reproduce the sound, respectively.

``Effects of the Invention'' As explained above, in this invention, the contact sound of a hard substance and the conversation sound are distinguished using the difference in their rising waveforms, and only the contact sound is suppressed. It is possible to suppress the most annoying sounds for users, and reduce the pain of the user.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG. 2 is a diagram showing examples of signal waveforms of various parts in the embodiment of FIG. 1. Patent applicant Nippon Telegraph and Telephone Corporation

Claims (1)

[Claims] (1) means for collecting sounds, means for extracting an envelope of the collected sounds, means for determining the magnitude relationship between the rise time of the extracted envelope and a predetermined reference time, and said collection. means for delaying the delayed sound for more than a reference time; means for erasing a part of the delayed sound when the judgment result is small according to the judgment result of the magnitude relationship; and reproduction of the sound with the part of the sound erased. A hearing aid consisting of a means for