JPH1070790A - Speaking speed detecting method, speaking speed converting means, and hearing aid with speaking speed converting function - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the hearing aid with the speaking speed converting function which can convert a speech signal to a proper speaking speed and output it whatever voicing speed (speaking speed) the speech signal has. SOLUTION: When the speech signal is inputted, the length Lpth of the vowel of the syllable at the head is measured. For example, when 'ohayou'(Good morning in English) is inputted, 'o' is inputted first, so when this 'o' is detected, its length is measured. On the basis of the length of the 'o', the speaking speed of the input speech signal is detected and a conversion rate for the speaking speed is calculated from the detected length and a target speaking speed value. Then 'hayou' inputted following the 'o' is converted with this conversion rate to perform speaking speed conversion corresponding to the speaking speed of the input speech signal almost in real time, so that the speech signal can be outputted at the target speaking speed all the times.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a hearing aid with a speech speed conversion function that compensates for a decrease in the hearing function of a wearer by extending the speech speed (speech speed) of an input voice signal and outputting the speech signal.

[0002]

2. Description of the Related Art Conventionally, hearing aids have been used as function assisting devices worn by persons with reduced hearing functions, such as the elderly. By the way, the deterioration of the hearing function due to aging is caused by a decrease in the sound transmission system function such as an increase in the minimum audible signal level and a decrease in the listening function in the high-frequency range, as well as the critical speed for speech identification (the maximum speech speed at which speech can be identified. ), Such as impaired auditory central function.

For this reason, as a hearing aid for the elderly, in addition to amplifying a part or all of the frequency band by temporally extending the audio signal, the output speed of the audio signal is made lower than the input speed. Hearing aids that perform speech rate conversion processing have also been proposed.

[0004]

However, in a hearing aid which merely converts an input voice signal at a low speed and outputs the converted signal, even if the interlocutor speaks slowly, the converted signal is output at a lower speed. As a result, the speech speed becomes too low, and even if the wearer is an elderly person, it may be difficult to hear.

To cope with this, it is only necessary to change the conversion rate of the speech rate conversion according to the speech rate of the speech of the interlocutor, but it is almost impossible for the elderly to perform this manually. Also, it is impossible to predict the utterance speed of the speaker in advance and determine the conversion rate.

According to the present invention, a speech speed detection method for measuring a speech speed at a head portion of an audio signal, and converting the speech speed of a subsequent audio signal into a target value using the detected speech speed, thereby real-time processing. Speech rate conversion method that enables speech rate conversion of a speech rate, and speech rate conversion that can convert a speech signal of any speech rate (speech rate) into an appropriate speech rate and output the speech signal It is an object to provide a hearing aid with a function.

[0007]

According to a first aspect of the present invention, the length of a first vowel in an input voice signal is detected, and the length of the input vowel is determined based on the length of the detected vowel. Detecting the speech speed of the voice signal.

In the invention of claim 2 of the present application, a target speech speed is set in advance, a speech signal is inputted, the speech speed of this speech signal is detected, and the speech signal detected by the inputted speech signal is detected. The speed is converted to the target speech speed.

According to a third aspect of the present invention, the length of the first vowel in the input voice signal is detected, and the speech rate of the input voice signal is determined based on the length of the detected vowel. And converting the speech rate of the voice signal input after the first vowel into a preset target speech rate based on the detected speech rate.

According to a fourth aspect of the present invention, there is provided an input means for inputting an audio signal including a voice signal, a voice speed detecting means for detecting a voice speed of the voice signal input from the input means, and a voice speed conversion. Target speech rate storage means for storing a target speech rate which is a target value of the speech rate, speech rate setting means for setting the target speech rate in the target speech rate storage means, and speech rate of the voice signal detected by the speech rate detection means. Conversion ratio calculating means for calculating a conversion ratio for converting from the first vowel to the target speech speed, and a speech for converting the speech speed of the voice signal inputted after the first vowel at the conversion ratio calculated by the conversion ratio calculating means. Speed conversion means.

According to a fifth aspect of the present invention, there is provided an input means for inputting an audio signal including an audio signal, an audio signal detecting means for monitoring the audio signal input from the input means and detecting the start of the audio signal. Vowel length detecting means for detecting the length of the first vowel of the voice signal whose start has been detected, and detecting the utterance speed of the voice signal whose start has been detected based on the length of the detected vowel. Speech speed detection means, target speech speed storage means for storing a target speech speed which is a target value of speech speed conversion,
A conversion ratio calculating unit for calculating a conversion ratio for converting the speech speed of the voice signal detected by the speech speed detecting unit to the target voice speed, and a conversion ratio calculated by the conversion ratio calculating unit after the first vowel. And a speech speed conversion means for converting the speech speed of the voice signal input to the communication device.

1 mora (1 syllable) for normal-rate utterances
Is about 140 to 150 ms. Further, since the consonant part and the vowel part overlap, it is difficult to specify the consonant part exactly.
It is known that s is occupied by consonants and vowels occupy 100-130 ms. It is also known that in ordinary conversations and announcements, the utterance speed does not change significantly with the utterance of about one word.

According to the first, third and fifth aspects of the present invention, based on these assumptions, the length of the first vowel of the voice signal is detected, and the temporal occupancy is calculated back to thereby make the utterance speed ( Speech speed) is detected. This makes it possible to detect the speech speed of the audio signal in real time (within about 200 ms) when the audio signal is input.

Further, in order to make it easier for the elderly to hear, it is preferable to extend one mora to about 200 ms (5 mora / sec). Claims 2, 4 and 5
In the invention of the above, the speech speed is set as a target speech speed, and the speech signal input so as to compensate for the difference between the speech speed detected by the above detection method and this target speech speed is converted into a desired speech speed. Even when speaking at a high speed, it is now possible to output an audio signal with a speaking speed that is easy for the elderly to hear.

[0015]

FIG. 1 is a block diagram showing the configuration of a hearing aid with a speech speed conversion function (hereinafter simply referred to as a hearing aid) according to an embodiment of the present invention. The microphone 10 receives the audio signal and inputs the audio signal to the amplifier 11. In addition,
The audio signal is an audible frequency signal composed of a voice signal or a noise, which is a human voice of a conversation or an announcement. Further, the microphone 10 may be provided anywhere such as a hearing aid body, an earpiece, or the like. The amplifier 11 amplifies the audio signal and inputs the amplified signal to the filter 12. Filter 1 above
Reference numeral 2 denotes an anti-aliasing filter, which is configured by a low-pass filter that cuts a frequency equal to or more than 1/2 of the sampling frequency. The audio signal passing through the filter 12 is converted into a digital signal (waveform data) by an A / D converter 13. The digital waveform data is input to the DSP 14. A signal processing RAM 15 and a parameter RAM 16 are connected to the DSP 14. The signal processing RAM 15 is a large-capacity one composed of a DRAM. The signal processing RAM 15 stores voice signals that have been converted in speech speed and expanded, and voice signals that are output after being delayed. The parameter RAM 16 stores D
This is a RAM for storing parameters for controlling the operation of the SP 14, and is configured by a battery-backed SRAM. A target speech speed data storage area 16a is set in the parameter RAM 16, and the number of expanded syllables (Nvmax), level threshold (Pth),
Parameters such as a length threshold (Lpth) and a limit wave number (Nd) are stored. A setting device 21 is connected to the parameter RAM 16. The setting unit 21 is for setting the target speech speed data and the number of expanded syllables.

The DSP 14 analyzes the input waveform data and determines whether or not a voice signal is currently input. If an audio signal is input, the signal is subjected to appropriate processing such as decompression and written to the signal processing RAM 15, and the signal written in synchronization with the read clock (sampling clock) is converted to a D / A converter. 1
7 is output. If no audio signal is input, the input signal is used as it is in the D / A converter 17.
Output to The DSP 14 is a D / A converter 17
When the signal is output to the, the equalizing for increasing the gain of the high frequency component of the signal is performed at the same time.

The D / A converter 17 reconverts the input digital waveform data into an analog audio signal and inputs the analog audio signal to a low-pass filter 18. The audio signal passes through the low-pass filter 18 to remove discontinuous noise during analog conversion. And amplifier 1
9 amplifies this audio signal to a level that can be heard by the user and outputs it to the receiver 20. Receiver 20
Converts the analog signal input from the amplifier 19 into air vibration and emits it to the ear canal of the wearer.

The A / D converter 13, DSP 14
The D / A converter 17 is supplied with a clock from a clock circuit (not shown).

Here, the function of the hearing aid will be described with reference to FIGS. The target speech speed data indicating the target value of the speech speed of the audio signal output from the receiver 20 is set to a parameter R
It is stored in the target speech speed data storage area 16a of the AM 16. The target speech speed data is set and input from the setting device 21. This setting may be performed at the time of factory shipment, or may be set by a user (wearer) himself. Regardless of the voice speed of the input voice signal, the voice speed is converted so that the voice signal is output at this voice speed. In order to perform the speech speed conversion processing according to the input voice in real time, the length of a vowel is measured without converting the voice speed of the first syllable of the input voice signal. The speech speed of the subsequent conversation is estimated from the length of the vowel. A speech speed conversion ratio (vowel waveform data expansion ratio) is calculated from the estimated speech speed of the input voice and the target value, and expansion processing is performed on the vowel of the following syllable. Thus, no matter what speed the speaker speaks, this sound signal is input to the wearer at a constant, most audible speech speed.

However, if all of the input voice signals are converted to speech speed and decompressed, if a long sentence is input at a time, the delay of the output voice signal with respect to the input voice signal becomes too large, so that the signal processing signal is too long. There is a problem in that the storage capacity of the RAM 15 becomes insufficient, the response shift in the conversation becomes too large, and the conversation is not performed smoothly.
In a television, a movie, or the like, there is a problem that a gap between the screen and the sound becomes large. On the other hand, even if a word or a series of sentences cannot be completely heard, if the part, especially the head part, can be accurately heard, the contents can often be sufficiently understood. . Therefore, attention is paid to the fact that the number of syllables constituting one word is relatively large, such as three or four, and in this embodiment, the first four syllables (three from the second syllable) are used. Is expanded, and the syllables thereafter are output as they are or compressed, thereby minimizing the delay between the input voice and the output voice. In this embodiment, in the case of Japanese, attention is paid to the fact that each syllable always includes a vowel, and the number of vowels is replaced by counting the number of syllables. Also, a part of the unvoiced section is deleted as necessary.

FIG. 2 is a diagram showing speech speed conversion processing when a four-syllable word "Ohayo" is input. The first "O"
Is input, it is output as it is without decompression, and the length of the vowel part is measured. The extension ratio is determined based on this length. When the next "ha" is entered, "ha"
The consonant H among the phonemes “Consonant: H” and “Vowel: A” constituting the same is output as is (memory (signal processing RAM 15)
The vowel A is expanded according to the expansion ratio and stored in the memory. The waveform data stored in the memory is sequentially read out by a reading program and output as an audio signal. The decompression processing program and the read program operate in parallel. Similarly, when the next "yo" is input, the consonant Y is stored in the memory as it is, and the vowel O is expanded and stored in the memory. Since "U" is only vowels, the whole is expanded and stored in the memory.

When a signal whose input level exceeds the input level threshold value Pth is longer than the duration threshold value Lpth, the hearing aid determines that the signal is an audio signal and performs the above processing. For this reason, even if the pulse-like noise shown in the figure is input, it is not processed as an audio signal because the duration is short. Also, the input audio signal (background sound such as level noise and pulse noise) is output as it is, except during the period in which the voice signal whose speech speed has been converted is read out and output from the memory.

Although this drawing shows only the expansion method on the time axis, it is assumed that the output level is actually amplified by several tens of dB compared to the input level. Further, this amplification level is not uniform in all frequency bands, but is limited to audio frequencies, and is amplified with a particularly large gain above audio frequencies. The DSP 14 also performs this equalizing process.

FIG. 3 is a diagram showing a speech speed conversion process when a nine-syllable sentence "Good morning" is input. The same processing is performed for the first four syllables “Ohayo” even for nine syllables, and the expanded audio signal is output from the receiver 20. Then, after the fifth syllable, it is compressed and output. The DSP 14 counts the number of syllables while performing decompression processing. If a syllable (vowel) continuously input after the fifth syllable is input is longer than a certain length, the DSP 14 counts this number. I try to compress.
After the fifth syllable, the wave number (cycle number) Nw of the vowel is counted, and when this count value Nw exceeds the limit number Nd, the subsequent numbers are compressed. The compression method reads two waves (two periods of waveform data), calculates an average wave of these waves, and stores only this one wave in a memory (the signal processing RAM 15) to compress the time to half. It is a method.

In this embodiment, the syllables after 5 syllables are compressed, but they may be output without compression. Further, as a compression method, when a syllable (vowel) exceeds a certain wave number, a method of compressing the excess portion to half is adopted. However, the compression method is not limited to this. The entire vowel may be compressed, or the non-compression limit may be determined in units of time instead of units of wave numbers.

FIGS. 4 to 8 are flowcharts showing the operation of the DSP. FIG. 4 is a data acquisition process, FIGS.
8 shows a speech speed conversion process, and FIG. 8 shows a reading process. These data fetching process, speech speed conversion process, and reading process are executed in parallel. It is assumed that the initial setting operation is executed prior to the start of all operations, and the signal processing RAM 15 is cleared, flags are preset, and the like.

The data fetch process will be described with reference to the flowchart of FIG. This data capturing process is executed for each frame composed of several samples of waveform data.
First, the waveform data D is fetched from the A / D converter 13 into a real-time buffer (s1). Then, this level is determined (s3). If the level of the data D is higher than the input level threshold Pth, the operation proceeds to s6 and below. If the level of D is equal to or lower than Pth, the process proceeds to s4 or lower. The real-time buffer and various flags are D
It is set inside SP14.

If there is no audio signal immediately after wearing, D ≦ Pt
h, Fns is set and Fs is reset by the initial setting, and the process returns without doing anything in s3 → s4 → s5. Here, the speech speed conversion flag Fs is a flag indicating that speech speed conversion processing (mainly expansion processing) is currently being performed. If this flag is set, it indicates that the input waveform data is not output as it is. I have. The signal absence flag Fns is a flag indicating whether or not the input waveform data exceeds Pth. If the reset state of the flag has continued for a predetermined time (Lpth) or more, that is, if a signal exceeding Pth has been continuously input for Lpth or more, it is determined that the input signal is an audio signal.

If any signal exceeding Pth is input, the process proceeds to s6, where it is determined whether or not the speech speed conversion flag Fs is set. At first, since this flag is not set, the process proceeds to s7 to determine whether or not the signal absence flag Fns is set. When the process first proceeds to this operation, the process proceeds from s7 to s8 because Fns is set. In s8, Fns is reset, and the timer counter T is reset to 0 in order to count the reset continuation time (time exceeding the threshold level) (s9). If the operation has progressed from s6 to s7 twice or more consecutively, F
Since ns has been reset, the process proceeds from s7 to s10.
In s10, 1 is added to the timer counter T. When the addition result T becomes equal to the threshold Lpth (s1
1) Assuming that the currently input signal is a voice signal, the speech speed conversion flag Fs is set to start the speech speed conversion process (s12), and the conversion ratio calculation process (FIG. 5) is started (s13). On the other hand, if T <Lpth in s11, the routine returns. As described above, when the waveform data D input in s2 has exceeded the level threshold Pth for a certain period of time Lpth or more, Fs is set because an audio signal has been input. While Fs is set,
Speaking speed conversion processing (conversion ratio calculation processing and conversion processing) to be described later is executed.

On the other hand, when the level D of the input waveform data falls below Pth, the process proceeds from s3 to s4. s
In step 4, it is determined whether or not Fns is set. After the level D once exceeds Pth, the level decreases and s
When the process proceeds to step S4, the process proceeds from step s4 to step s14 because Fns is reset in step s8. At s14, the signal absence flag Fns is set, and the timer counter T is reset to count the silence time (s15). If Fns has already been reset, s4 to s5
To determine whether or not the speech speed conversion flag Fs has been reset. When Fs is reset, the process returns as described above. However, once the speech speed conversion operation is started, the input signal level is reduced, and if the process proceeds to this processing operation, Fs remains set. Because there is, the process proceeds from s5 to s16. In s16, 1 is added to the timer counter T. When the addition result T reaches the silence time threshold Tns (s17), it is determined that the input of the audio signal has already been completed, the speech speed conversion flag Fs is reset, and the waveform data of the audio signal ends. After that, an instruction is given to the speech speed conversion processing operation which is operating in parallel so as to discard the data of the silent portion written in the signal processing RAM 15 (s19). On the other hand, even if 1 is added to T, T <T in s17.
If it is Tns, the process returns.

As described above, when the input waveform data is Pt
When Tns has elapsed while h is still lower than h, the speech speed conversion flag Fs is reset assuming that the input of the audio signal has ended. The reason why Fs is not reset immediately when the level D of the waveform data falls below Pth is that there is a short-time silent part (voiceless section) in the voice signal, and this voiceless section is also used as the voice signal. It is necessary to take in. The silent portion included in the audio signal includes a prompting sound "tsu" and an interval between words. Hereinafter, the speech speed conversion processing operation will be described. The flowchart of FIG. 5 shows the conversion ratio calculation processing operation that is executed first when Fs is set and the speech speed conversion processing operation starts. When this operation starts, since the waveform data for Lpth is accumulated in the real-time buffer, an appropriate section is cut out (s21), and the fundamental frequency fz of each section is calculated based on the interval between the zero cross points (s22). . A vowel part is extracted based on the fz (s23). Since the fundamental frequency of the vowel part is lower than that of the consonant part, these can be separated and extracted. And the vowel part number counter Nv
Is set to 1 (s24). This vowel part number counter N
v counts the number of vowel parts instead of counting the number of syllables. In the following processing, this count value is treated as the number of syllables. Hereinafter, while monitoring the waveform data input to the real-time buffer, the temporal length Lv of the vowel is counted until the vowel portion is completed (s25, s).
26). When the vowel part of this syllable is completed (s26), the length of the syllable is estimated based on the length of this vowel part, and the speech speed (speech speed) of this voice signal is calculated based on this (s27). The calculated speech speed and parameter RAM1
The speech speed conversion ratio is calculated by comparing the target speech speed data stored in No. 6 with the target speech speed data (s28). Thereafter, a conversion processing operation (FIG. 6) for executing the speech speed conversion processing is started (s29).

The real-time buffer can store and store waveform data for a certain period of time, and the processed data is appropriately cleared or overwritten in each processing operation.

FIG. 6 is a flowchart showing the conversion processing operation for actually executing the speech speed conversion processing. This operation is executed from the data of the second syllable after the audio signal is input. When this processing is started, the input waveform data cannot be directly read out and output from the real-time buffer and must be read out from the signal processing RAM 15, so that the memory read flag Fm is set (s3).
0). Then, the data stored in the real-time buffer is read (s31). Whether this data is vowel part data (s32) or consonant part data (s3)
3) It is determined whether the data is data of a silent part (s34) or data of an instruction to discard the silent part data (s35). In this operation, the reading of the data of the real-time buffer is not performed only in s31, but is performed in each processing operation as needed. Further, when reading data from the real-time buffer, it waits until data is input from the A / D converter 13 as necessary.

Since the syllables other than the row start from a consonant,
If it is determined that the consonant part is present, the process proceeds from s33 to s36, and the data of this consonant part is written as it is to the signal processing RAM 15 (s36). This is because consonants are non-periodic sounds and become unnatural when processed, so they are not expanded even when speech speed conversion is performed.

On the other hand, if the read data is vowel part data, 1 is added to the vowel part number counter Nv (s40). Thus, it is determined whether Nv has exceeded Nvmax (s41). As described with reference to FIG. 3, in this embodiment, Nvmax = 4. Therefore, when Nv becomes 5, s41 becomes a positive determination and proceeds to s45. When Nv is equal to or less than Nvmax,
This vowel part is extended (s42, s43).

The decompression process will be described with reference to the flowchart of FIG. Here, a plurality of waves of the vowel part are treated as one block. For example, one block is composed of three vowel waves. Then, the fundamental frequency of the vowel waveform in this block is calculated (s60). The calculation of the fundamental frequency uses a zero cross, which is almost the same as the operation in s22. Then, two adjacent waveforms in the block are selected and cut out (s61), and their average waveform is calculated (s61).
62). Then, this average waveform is inserted between the two cut-out waveforms (s63). The block now has four waves. This four-wave block is used for signal processing RA.
Write to M15 (s64). In this example, since three waves are expanded into four waves, the expansion ratio is 133%. Also,
Each block does not need to have the same number, and the decompression rate is 130
In order to obtain the percentage, the wave number of the block may be repeated 4, 3, and 3.

Returning to FIG. 6, detected vowel parts (syllables)
Is the fifth or later, s41 to s45
Proceed to. In s45, the wave number counter Nw for measuring the length of the vowel portion by the wave number is cleared. Then, the waveform data input to the real-time buffer is read, and Nw is counted up each time one waveform is input (s4).
6). Until the wave number Nw exceeds the limit wave number Nd, the data is directly written to the RAM (s47 → s4
8) When Nw exceeds the limit wave number Nd, compression processing is performed thereafter (s49), and the result is written into the signal processing RAM 15.

FIG. 7B is a flowchart showing the compression processing operation. This operation has two
It is executed after a waveform is input. First, this 2
The waveform is cut out (s65), and the average waveform is calculated (s66). Then, the calculated average waveform is referred to the above 2
Write to the signal processing RAM 15 instead of the waveform (s6
7). By this operation, the vowel part waveform after Nd becomes １ ／.
Will be compressed.

If the waveform data read in s31 is for a silent part, the value of the vowel number (syllable number) counter Nv is determined (s55). If it does not exceed Nvmax, the value is adjusted to the expansion rate of the vowel part. The silent part of the lever is also expanded and written into the signal processing RAM 15 (s56). If the number of syllables N
If v exceeds the expansion limit number Nvmax, it is written into the signal processing RAM 15 without expansion (s3
6). On the other hand, if the read data is an instruction to discard a silent part, the silent data group stored at the end of the signal processing RAM 15 is discarded / erased (s57). This is because these silence part data was stored as unvoiced sections (prompting sounds, etc.) of the audio signal, but it was actually found to be a silence part after the end of the audio signal and was unnecessary. When the discard instruction is input, it means that the processing of the audio signal has been completed, so that this operation is completed and the process returns.

FIG. 8 is a flowchart showing the read processing operation. This operation is started and performed at the same time as the operation of the hearing aid as in the data acquisition process. This operation is also executed at each sampling timing, similarly to the data fetch processing operation.

First, it is determined whether or not Fm is set (s70). If Fm is not set, s7
In step 4, the latest data stored in the real-time buffer is read and output to the D / A converter 17. RAM for signal processing when Fm is set
15 to determine whether there is data to be read (s7).
1) In some cases, the data at the position indicated by the time pointer is read and output to the D / A converter 17 (s)
72). The time pointer is incremented by the reading, but it may be changed in some cases by data writing by the conversion processing (including the decompression processing and the compression processing). On the other hand, when there is no data to be read in the signal processing RAM 15, after resetting Fm (s7
3) The latest data is read from the real-time buffer and output to the D / A converter 17 (s74). Thereafter, data is output from the real-time buffer to the D / A converter 17 until the conversion processing operation is started and Fm is set. The signal processing RAM
At 15, it is assumed that an operation of erasing read-out data is appropriately performed.

In the above embodiment, the first vowel (syllable) is output without converting the speech speed. However, the output may be converted at a certain conversion ratio. For example, the conversion is performed at a conversion ratio determined for the immediately preceding audio signal.

In the above embodiment, the speech speed of the voice signal is detected based on the length of the first vowel, but the method of detecting the speech speed is not limited to this. For example, when an audio signal is input after a silent part, a method of estimating the utterance speed of the current audio signal from the vowel length of the audio signal input immediately before the silent part, or when an audio signal is input after the silent part Then, a method of estimating the utterance speed of the current voice signal from the distance between vowels of the voice signal input immediately before the silent portion can be adopted. According to these methods, the immediately preceding syllable can be used, and the vowel length and the distance between vowels of the immediately preceding sentence are obtained, and the average value and its change curve are used to estimate the current utterance speed. Can be.

Further, this embodiment includes the following inventions not described in the claims.

By preventing the speech speed from being converted after a predetermined syllable, it is possible to prevent a decrease in the degree of comprehension and to minimize the output delay.

The intelligibility can be increased by increasing the gain of the signal detected as the audio signal.

[0047]

As described above, according to the present invention, the utterance speed of an audio signal is detected based on the first vowel of the audio signal, thereby detecting the utterance speed with high accuracy in almost real time. be able to.

According to the present invention, the speech rate of the input speech signal is detected, and the speech rate of the speech signal is converted into the target speech rate, whereby the speech signal of any speech rate is inputted. In this case, the user can convert the speech speed to a desired speed (target speech speed) in real time.

[Brief description of the drawings]

FIG. 1 is a block diagram of a hearing aid with a speech speed conversion function according to an embodiment of the present invention;

FIG. 2 is a diagram illustrating a speech speed conversion function of the hearing aid.

FIG. 3 is a diagram illustrating a speech speed conversion function of the hearing aid.

FIG. 4 is a flowchart showing the operation of the DSP of the hearing aid;

FIG. 5 is a flowchart showing the operation of the DSP of the hearing aid;

FIG. 6 is a flowchart showing the operation of the DSP of the hearing aid;

FIG. 7 is a flowchart showing the operation of the DSP of the hearing aid;

FIG. 8 is a flowchart showing the operation of the DSP of the hearing aid.

[Explanation of symbols]

10 microphone, 11 microphone amplifier, 12 filter, 13 A / D converter, 14 DSP, 15
... Sound signal RAM, 16 ... Parameter RAM, 16a ...
Target speech speed data storage area, 17 ... D / A converter,
18 low-pass filter, 19 power amplifier, 20
Receiver, 21 ... Setting device

Claims (5)

[Claims] 1. The method according to claim 1, wherein a length of a first vowel in the input voice signal is detected, and an utterance speed of the input voice signal is detected based on the length of the detected vowel. Talk speed detection method. 2. A target speech speed is set in advance, an audio signal is input, an utterance speed of the audio signal is detected, and the input speech signal is converted from the detected utterance speed to the target speech speed. A speech speed conversion method characterized by converting. Detecting a length of a first vowel in the input voice signal; detecting a speech rate of the input voice signal based on the length of the detected vowel; A speech speed conversion method, comprising: converting the speech speed of a speech signal input after the first vowel based on the speech speed to a preset target speech speed. 4. An input means for inputting an acoustic signal including a voice signal, a voice speed detecting means for detecting a voice speed of the voice signal input from the input means, and a target voice speed which is a target value of the voice speed conversion. Speech speed setting means for setting the target speech speed in the target speech speed storage means, and converting the speech speed of the voice signal detected by the speech speed detection means into the target speech speed. Conversion rate calculating means for calculating a conversion rate for the conversion rate; and speech rate conversion means for converting the speech rate of a voice signal input after the first vowel with the conversion rate calculated by the conversion rate calculating means. A hearing aid with a speech speed conversion function. 5. An input means for inputting an audio signal including an audio signal, an audio signal detecting means for monitoring the audio signal input from the input means and detecting the start of the audio signal, and detecting the start of the audio signal. Vowel length detection means for detecting the length of the first vowel of the voice signal; speech speed detection means for detecting the utterance speed of the voice signal whose start has been detected based on the length of the detected vowel; A target speech speed storage unit that stores a target speech speed that is a target value of the speed conversion; and a conversion ratio that calculates a conversion ratio for converting the speech speed of the voice signal detected by the speech speed detection unit into the target speech speed. Speech rate conversion means for calculating speech rate of speech signals input after the first vowel at the conversion rate calculated by the conversion rate calculation means. Hearing aid with function.