JP3034494B2 - Interactive speech rate conversion apparatus and method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and a method for interactive speech rate conversion for performing speech rate conversion, and more particularly to a speech rate conversion hearing aid, a speech rate conversion telephone, and a speech rate conversion apparatus for the hearing impaired and the elderly. The present invention relates to a real-time interactive speech rate conversion device and method for improving audibility when performing auxiliary listening by speech rate conversion in an interactive communication means such as a communication device.

[Summary of the Invention] The present invention relates to an interactive communication means such as a speech rate conversion hearing aid, a speech rate conversion telephone, or a speech rate conversion communication device, which expands the length of speech to increase the speed of utterance. When using the speech speed conversion function that changes the (speech speed) to "slow", the listener informs the speaker that the speech speed conversion function is to be used, and the speaker finishes uttering and waits for a response from the listener. In the meantime, the listener using the speech rate conversion function using various methods informs the speaker of the timing of listening to the speech rate converted voice, eliminating the discomfort caused by the response delay of the listener side And
It enables smooth dialogue.

[0003]

2. Description of the Related Art Conventionally, when a speech speed conversion function is used for an interactive communication means such as a hearing aid, a telephone, or a communication device, even if the speaker has already finished speaking,
For listeners listening using the speech speed conversion function,
Since the utterance of the speaker remains as much as the speech speed changes slowly, the response timing of the listener becomes slower than in a normal conversation.

[0004] For this reason, there is a problem that it becomes difficult for the speaker side to take the timing of the dialogue because the response of the listener is slower than in the normal dialogue, and it becomes difficult to perform a smooth dialogue.

In order to solve such a problem, Japanese Patent Laid-Open Publication No. Hei 6-311211 proposes a "voice speed conversion telephone and a voice speed conversion adapter".

According to this proposal, when the speech speed conversion function is used by the telephone, while the speaker is speaking,
The voice that the listener is listening to has become slower,
Feedback to the speaker at a low volume to inform the speaker of the timing of the sound that the listener is listening to, so that the speaker can grasp the timing of the listener's response, so that a smooth dialogue can be performed. I have.

[0007]

However, when the proposed method is used for interactive communication means, no matter how low the volume is, the speaker must hear the sound heard by the listener. However, the utterance of the speaker is disturbed, and the utterance of the speaker is made difficult, so that a smooth dialogue becomes impossible.

For this reason, conventionally, a speech speed conversion function is provided by a hearing aid,
It is practically impossible to use it for interactive communication means such as a telephone and a communication device, and there is a problem that speech speed conversion cannot be performed by the interactive communication means.

SUMMARY OF THE INVENTION In view of the above circumstances, the present invention makes it possible to recognize that the response timing of the listener is delayed, to eliminate the sense of incongruity caused by the response delay, and to allow the speaker to adjust the utterance timing so that it can be used during conversation. It is a main object of the present invention to provide an interactive speech speed conversion device and method that can eliminate the discomfort of the user.

[0010]

[0011]

SUMMARY OF THE INVENTION To achieve the above object, according to the present invention, there is provided an interactive speech speed conversion apparatus for converting a speech speed spoken by a speaker and listening. During a conversation using, a conversation interruption detector that detects a state in which the speaker's utterance ends and is waiting for a response from the listener, and a conversation interruption is detected by the conversation interruption detector. And a notifying unit for notifying the talker of the talk speed converted voice heard by the listener as it is or by reducing the volume.

According to a second aspect of the present invention, in the interactive speech speed conversion device for converting the speech speed uttered by the speaker and listening, the speech of the speaker is terminated when the conversation is performed using the speech speed conversion function. A conversation interruption detecting unit for detecting a state of waiting for a response from the listener, and a low-frequency component of a speech speed converted speech heard by the listener when the conversation interruption is detected by the conversation interruption detecting unit. And a notifying unit for notifying the speaker at the same volume or a reduced volume after the extraction.

According to a third aspect of the present invention, in the interactive speech speed conversion device for converting the speech speed spoken by the speaker and listening, the speech of the speaker is terminated when the conversation is performed using the speech speed conversion function. , A conversation interruption detection unit that detects the state of waiting for the response of the listener, and changes the characteristics of the speech speed converted voice that the listener is listening to when the conversation interruption is detected by the conversation interruption detection unit And a notifying unit for notifying the speaker of the same volume or the reduced volume after the completion.

According to a fourth aspect of the present invention, in the interactive speech speed conversion device for converting the speech speed uttered by the speaker and listening, the speech of the speaker is terminated when the conversation is performed using the speech speed conversion function. A conversation interruption detecting unit that detects a state of waiting for a response from the listener, and that when the conversation interruption is detected by the conversation interruption detecting unit, the listener is listening to the voice whose speech speed has been converted. And a notifying unit for notifying the speaker of the voice message or signal shown.

According to a fifth aspect of the present invention, in the interactive speech speed conversion device according to any one of the first to fourth aspects, when the listener notifies the speaker that the listener is listening to the voice whose speech speed has been converted, the listener can perform the conversion. It is characterized in that the uttered voice is sent to the speaker.

According to a sixth aspect of the present invention, in the interactive speech speed conversion device according to any one of the first to fifth aspects, the conversation interruption detecting unit and the notifying unit include a speech speed conversion telephone on the listener side and a speech on the listener side. It is characterized in that it is provided in either a speed conversion communication device or a speech speed conversion hearing aid on the listener's side.

[0017]

According to a seventh aspect of the present invention, in the interactive speech speed conversion method of converting the speech speed spoken by the speaker and listening, the speech of the speaker is terminated when the conversation is performed using the speech speed conversion function. , Detects the state waiting for the response of the listener,
The speech speed converted voice that the listener is listening to is
Alternatively, the speaker is notified by reducing the volume.

According to the present invention, in the interactive speech speed conversion method of converting the speech speed spoken by the speaker and listening, the speech of the speaker is terminated when the conversation is performed using the speech speed conversion function. Detecting the state of waiting for the response of the listener, extracting the low-frequency component of the speech-rate-converted voice heard by the listener, and then notifying the speaker at the same volume or at a reduced volume. It is characterized by.

According to a ninth aspect of the present invention, in the interactive speech speed conversion method of converting the speech speed spoken by the speaker and listening, the speech of the speaker is terminated when the conversation is performed using the speech speed conversion function. Detects the state of waiting for the response from the listener, changes the characteristics of the voice converted by the listener, and then notifies the speaker at the same volume or at a reduced volume. And

According to a tenth aspect of the present invention, in the interactive voice speed conversion method of converting the voice speed spoken by the speaker and listening, the voice of the speaker is terminated when the voice is converted using the voice speed conversion function. , Waiting for a response from the listener,
It is characterized in that a voice message or a signal indicating that the listener is listening to the voice whose speech speed has been converted is notified to the speaker.

In the eleventh aspect, in the interactive speech speed conversion method according to any one of the seventh to tenth aspects, when the listener side notifies the speaker side that he / she is listening to the voice whose speech speed has been converted, the listener side It is characterized in that the uttered voice is sent to the speaker.

[0023]

[0024] In the above-mentioned configuration, in the first aspect, when a conversation is performed using the speech speed conversion function, when the utterance of the speaker is terminated and the response of the listener is awaited, the conversation interruption detecting unit is provided. And the notification unit notifies the speaker side of the voice that the listener has been listening to and that has undergone speech rate conversion as it is or by reducing the volume. In this way, the speaker's side utterance is not hindered, the listener's side response timing is grasped, the discomfort caused by the response delay is eliminated, and the speaker's utterance timing is adjusted to eliminate the discomfort during conversation.

According to a second aspect of the present invention, when the conversation is finished using the speech speed conversion function, the utterance of the speaker is terminated and the conversation is interrupted by the conversation interruption detecting unit when waiting for the response of the listener. After extracting the low-frequency component of the speech speed-converted sound that the listener is listening to by the notification unit,
Notify the speaker at the same or reduced volume. In this way, the speaker's side utterance is not hindered, the listener's side response timing is grasped, the discomfort caused by the response delay is eliminated, and the speaker's utterance timing is adjusted to eliminate the discomfort during conversation.

According to a third aspect of the present invention, when a conversation is performed using the speech speed conversion function, when the utterance of the speaker is terminated and a response from the listener is awaited, this is detected by the conversation interruption detecting unit. At the same time, after changing the characteristics of the speech-speed-converted voice that the listener is listening to, the notification unit notifies the speaker at the same volume or a reduced volume. In this way, the speaker's side utterance is not hindered, the listener's side response timing is grasped, the discomfort caused by the response delay is eliminated, and the speaker's utterance timing is adjusted to eliminate the discomfort during conversation.

According to a fourth aspect of the present invention, during a conversation using the speech speed conversion function, when the speaker's utterance ends and the listener's side waits for a response, this is detected by the conversation interruption detector. At the same time, the notification unit notifies the speaker of a voice message or a signal indicating that the listener is listening to the voice whose speech speed has been converted. In this way, the speaker's side utterance is not hindered, the listener's side response timing is grasped, the discomfort caused by the response delay is eliminated, and the speaker's utterance timing is adjusted to eliminate the discomfort during conversation.

According to a fifth aspect of the present invention, when a conversation is being performed between the speaker and the listener using the speech speed conversion function, while the speaker has finished speaking and is waiting for a response from the listener, Then, the speaker is notified of a voice or a signal indicating whether or not the listener is listening to the voice whose speech speed has been converted, and a voice uttered by the listener. In this way, the speaker's side utterance is not hindered, the listener's side response timing is grasped, the discomfort caused by the response delay is eliminated, and the speaker's utterance timing is adjusted to eliminate the discomfort during conversation.

According to a sixth aspect of the present invention, the conversation interruption detecting section and the notifying section are provided in any of a listener's side speech rate conversion telephone, a listener's side speech rate conversion communication device, and a listener's side speech rate conversion hearing aid. When talking between the talker and the listener using any of the talk-rate conversion telephone, the talk-rate conversion communicator, and the talk-rate conversion hearing aid, the talker finishes speaking and the listener While waiting for a response, the speaker is informed of the sound or signal indicating whether the listener is listening to the speed-changed sound, and if necessary, the sound uttered by the listener, so that the speaker can speak. While not disturbing, let the listener understand the response timing of the listener side, eliminate discomfort caused by the response delay,
Let the speaker adjust the utterance timing to eliminate discomfort during conversation.

[0030]

According to a seventh aspect of the present invention, when a conversation is performed between the speaker and the listener using the speech speed conversion function, while the speaker finishes speaking and waits for a response from the listener, Informing the speaker of the speed-converted voice that the listener is listening to, so that the listener's response timing can be grasped while not disturbing the speaker's utterance,
The discomfort caused by the response delay is eliminated, and the talker's side adjusts the utterance timing to eliminate the discomfort during conversation.

According to the eighth aspect, when a conversation is being performed between the speaker and the listener using the speech speed conversion function, while the speaker has finished speaking and is waiting for a response from the listener, , The low-frequency component of the speech-speed-converted voice heard by the listener is extracted and notified to the speaker, thereby preventing the speaker's utterance from being disturbed, allowing the listener to grasp the response timing, and delaying the response. And eliminate the discomfort during conversation by adjusting the utterance timing on the speaker side.

According to the ninth aspect, when a conversation is being performed between the speaker and the listener using the speech speed conversion function, while the speaker finishes speaking and waits for a response from the listener, , By changing the characteristics of the speech speed-converted speech that the listener is listening to, notifying the speaker, and thereby not obstructing the speaker's utterance, allowing the listener's response timing to be grasped, resulting in a response delay And let the speaker adjust the utterance timing,
Eliminate discomfort during conversation.

In the tenth aspect, when a conversation is performed between the speaker and the listener using the speech speed conversion function,
While the speaker has finished speaking and is waiting for a response from the listener, the listener is notified of a voice or signal indicating whether or not the listener is listening to the converted speech, thereby preventing the speaker from speaking. In such a case, the response timing on the listener side is grasped, the discomfort caused by the response delay is eliminated, and the speaker side adjusts the utterance timing to eliminate the discomfort during conversation.

In the eleventh aspect, when a conversation is performed between the speaker and the listener using the speech speed conversion function,
While the speaker has finished speaking and waiting for a response from the listener, inform the speaker of a voice or signal indicating whether or not the listener is listening to the speed-converted voice, and the voice of the listener. In this way, the speaker's side utterance is not hindered, the listener's side response timing is grasped, the discomfort caused by the response delay is eliminated, and the speaker's utterance timing is adjusted to eliminate the discomfort during conversation.

[0036]

DESCRIPTION OF THE PREFERRED EMBODIMENTS << Description of First Embodiment >> FIG.
FIG. 1 is a block diagram illustrating an example of an interactive communication system using a first embodiment of an interactive speech speed conversion device and method according to the present invention.

The interactive communication system 1a shown in FIG. 1 includes a telephone set 2 provided on the speaker side, a telephone set 3a provided on the listener side, and a line 4 connecting the telephone sets 2 and 3a. Phone 2
When a call is started when the listener's telephone 3a goes off-hook and the telephone is started, the listener's telephone 3a detects this, generates a voice message or signal, and transmits it to the speaker's telephone 2. Then, the speaker is notified that the listener uses the speech speed conversion function, and the speaker recognizes that the response of the listener is delayed, so that the conversation between the speaker and the listener is smoothly performed.

In this case, the telephone 3 provided on the listener side
a, when a talker's side call signal is supplied from the line 4 as shown in FIG. 2, or as shown in FIG.
The speaker calls the speaker from the listener and the speaker's telephone 2
Is off-hook and a reception signal is supplied from the line 4, the reception signal is received by the reception detection unit 5, and
Based on the detection result of the reception detection unit 5, the message signal transmission unit 6 is activated, and the listener uses the speech speed conversion function to make a call before the talker and the listener start talking. Is generated and transmitted to the telephone set 2 of the talker side to notify the talker side that the listener uses the speech speed conversion function.

This allows the speaker to recognize that the response of the listener is delayed, and prepares for the delay of the response in the dialogue, thereby smoothing the conversation between the speaker and the listener.

As described above, in this embodiment, when the talker and the listener start talking, this is detected by the reception detector 5 provided in the telephone 3a of the listener and the message signal is transmitted. The transmitting unit 6 is activated to generate a voice message or a signal indicating that the listener uses the speech speed conversion function to make a call, and transmits the voice message or signal to the speaker's telephone 2 so that the listener can talk to the speaker. Since the notification that the speed conversion function is to be used is made, the speaker can recognize that the response timing of the listener is delayed, and the discomfort caused by the response delay can be eliminated.
The speaker can adjust the utterance timing to eliminate discomfort during conversation.

<< Description of Second Embodiment >> FIG. 4 is a block diagram showing an example of an interactive communication system using an interactive speech speed conversion apparatus and method according to a second embodiment of the present invention. In this figure, the same parts as those in FIG. 1 are denoted by the same reference numerals.

The interactive communication system 1b shown in this figure is different from the interactive communication system 1a shown in FIG. 1 in that the telephone 3a for transmitting a voice message or a signal at the start of a telephone call is replaced by the time when the talk of the talker is interrupted. Then, a telephone 3b for transmitting the voice heard by the listener is provided. When the voice of the speaker is analyzed by the telephone 3b and a pause section is detected, the voice heard by the listener (voice generated by the speaker is transmitted). The converted voice is fed back to the speaker,
The purpose is to make the listener recognize the voice being heard, and to make the call between the speaker and the listener smooth.

In this case, the listener's telephone 3b is
As shown in (1), while analyzing the voice from the speaker side for each short time block by the analysis processing unit 7, the characteristics of the voice are analyzed, and based on the analysis information output from the analysis processing unit 7, the speech speed conversion and synthesis processing is performed. By the part 8, (a) of FIG.
As shown in (b), the voice of the speaker is converted to the voice speed, the listener can hear the slow voice, and the volume of the voice output from the voice speed conversion / synthesis processing unit 8 is controlled by the volume control unit 9. Lower.

In parallel with this operation, based on the analysis information output from the analysis processing unit 7, the silence determination unit 10 determines, for each short-time block, whether or not the speaker's voice is a silence block. When several blocks in the past are determined to be silent blocks by the silent determining unit 10 and the current block is determined to be a silent block, the pause determining unit 11 determines that the current block includes a pause, Based on the determination result, the signal switching unit 13
In this way, the sound whose volume has been adjusted (the sound that the listener has listened to and whose speech speed has been converted) output from the volume controller 9 is selected, and this is fed back to the speaker as shown in FIG. Then, the speaker recognizes the voice heard by the listener.

Thus, when the speaker's utterance is interrupted, the listener is made to recognize which part of the slowly converted voice is currently being heard, and the next utterance timing is measured based on the recognition result. Communication between speakers and listeners.

When the silence judging unit 10 does not judge several past blocks as silence blocks, or when the current block is not judged as silence blocks,
The pause determination unit 11 determines that the current block does not include a pause. Based on the determination result, the signal switching unit 13 selects the zero signal output from the zero signal transmission unit 12 at all times. In this case, silence is fed back to the speaker so as not to disturb the utterance of the speaker.

As described above, in the second embodiment,
When the speaker's utterance continues, the speaker's telephone 2 is prohibited from feeding back the voice heard by the listener, so that the speaker's utterance is not interrupted, and the speaker's utterance is interrupted. At this point, the listener's telephone 3b detects this, feeds back the voice heard by the listener to the speaker at a low volume, and causes the speaker to recognize the voice heard by the listener. While not disturbing the speaker's utterance, the listener's response timing can be grasped and discomfort caused by the response delay can be eliminated, and the speaker's utterance timing can be adjusted to eliminate discomfort during conversation be able to.

<< Description of Third Embodiment >> FIG. 7 is a block diagram showing an example of an interactive communication system using an interactive speech speed conversion apparatus and method according to a third embodiment of the present invention. In this figure, the same parts as those in FIG. 5 are denoted by the same reference numerals.

The difference between the interactive communication system 1c shown in this figure and the interactive communication system 1b shown in FIG. 5 is that the structure of the telephone set 3b for transmitting the voice heard by the listener when the voice of the speaker is interrupted. Is partially provided, and a telephone 3c is provided.
c, the voice of the speaker is analyzed, and when a pause section is detected, the voice heard by the listener (the voice obtained by converting the voice generated by the speaker) is subjected to low-pass filtering (low-pass filtering). This is to provide feedback to the speaker side so that the speaker side recognizes the voice heard by the listener side, so that the call between the speaker and the listener can be smoothly performed.

In this case, the telephone 3c on the listener side is
As shown in (1), while analyzing the voice from the speaker side for each short time block by the analysis processing unit 7, the characteristics of the voice are analyzed, and based on the analysis information output from the analysis processing unit 7, the speech speed conversion and synthesis processing is performed. The voice of the speaker is converted into the voice by the unit 8 and the slow voice is heard by the listener, and the low-pass filtering unit 14 controls the low-frequency component of the voice output from the voice-speed conversion / synthesis unit 8. After extracting only
The volume of the sound output from the low-pass filter 14 is reduced by the volume controller 9.

In parallel with this operation, based on the analysis information output from the analysis processing unit 7, the silence determining unit 10 determines whether the speaker's voice is a silent block for each short-time block, When several blocks in the past are determined to be silent blocks by the silent determining unit 10 and the current block is determined to be a silent block, the pause determining unit 11 determines that the current block includes a pause, Based on the determination result, the signal switching unit 13
, The volume-adjusted sound output from the volume adjustment unit 9 (the sound that is heard by the listener and that has been subjected to low-frequency filtering of the speech-speed-converted sound) is selected, and this is fed back to the speaker to provide the speaker. Side recognizes the voice that the listener is listening to.

Thus, when the utterance of the speaker is interrupted, the listener is made to recognize which part of the slowly converted voice is currently being heard, and the next utterance timing is measured based on the recognition result. Communication between speakers and listeners.

When the silence judging section 10 judges that some past blocks have not been judged as silence blocks, or when the current block has not been judged as silence blocks,
The pause determination unit 11 determines that the current block does not include a pause. Based on the determination result, the signal switching unit 13 selects the zero signal output from the zero signal transmission unit 12 at all times. In this case, silence is fed back to the speaker so as not to disturb the utterance of the speaker.

As described above, in the third embodiment,
When the speaker's utterance continues, the speaker's telephone 2 is prohibited from feeding back the voice heard by the listener, so that the speaker's utterance is not interrupted, and the speaker's utterance is interrupted. At this point, this is detected by the listener's telephone 3c, and the phonological properties of the voice heard by the listener are reduced, and the voice is fed back to the speaker at a low volume, and the voice heard by the listener is recognized by the speaker. Because it does not disturb the speaker's utterance, it is possible to grasp the listener's response timing, eliminate the discomfort caused by the response delay, and adjust the speaker's utterance timing It is possible to eliminate discomfort during conversation.

<< Description of Fourth Embodiment >> FIG. 9 is a block diagram showing an example of an interactive communication system using an interactive speech speed conversion apparatus and method according to a fourth embodiment of the present invention. In this figure, the same parts as those in FIG. 5 are denoted by the same reference numerals.

The difference between the interactive communication system 1d shown in this figure and the interactive communication system 1b shown in FIG. 5 is that the configuration of the telephone set 3b for transmitting the voice heard by the listener when the voice of the speaker is interrupted. Is provided with a telephone 3d which is partially modified.
According to d, the speaker's voice is analyzed, and when a pause section is detected, the voice heard by the listener (the voice converted from the voice emitted by the speaker) is processed to change the characteristics of the speaker's voice. Then, feedback is provided to the speaker, and the speaker recognizes the voice heard by the listener, so that the conversation between the speaker and the listener is smoothly performed.

In this case, the listener's telephone 3d is
As shown in FIG. 0, while analyzing the speech from the speaker side for each short time block by the analysis processing unit 7, the characteristics of the voice are analyzed, and based on the analysis information output from the analysis processing unit 7, the speech speed conversion synthesis is performed. The processing unit 8 converts the voice of the speaker into a speech speed, and allows the listener to hear a slow voice, and the voice quality processing unit 15 outputs an average high voice output from the speech speed conversion / synthesis processing unit 8. Change the length,
After changing the speaker structure (personality) by changing the spectral structure of the voice or changing the temporal change of the pitch of the voice, the sound is output from the voice quality processing unit 15 by the volume control unit 9. Decrease audio volume.

In parallel with this operation, based on the analysis information output from the analysis processing unit 7, the silence determination unit 10 determines whether the speaker's voice is a silence block for each short time block, When several blocks in the past are determined to be silent blocks by the silent determining unit 10 and the current block is determined to be a silent block, the pause determining unit 11 determines that the current block includes a pause, Based on the determination result, the signal switching unit 13
, The volume-adjusted voice output from the volume control unit 9 (the voice that is heard by the listener and whose voice quality has been converted and converted into the voice quality) is selected and fed back to the speaker, and the speaker's side is selected. Then, the listener is made to recognize the sound being heard.

Thus, when the speaker's speech is interrupted, the listener is made to recognize which part of the slowly converted speech is currently being heard, and the next speech timing is measured based on the recognition result. Communication between speakers and listeners.

When the silence judging section 10 does not judge several past blocks as silence blocks, or when the current block is not judged as silence blocks,
The pause determination unit 11 determines that the current block does not include a pause. Based on the determination result, the signal switching unit 13 selects the zero signal output from the zero signal transmission unit 12 at all times. In this case, silence is fed back to the speaker so as not to disturb the utterance of the speaker.

As described above, in the fourth embodiment,
When the speaker's utterance continues, the speaker's telephone 2 is prohibited from feeding back the voice heard by the listener, so that the speaker's utterance is not interrupted, and the speaker's utterance is interrupted. At this point, this is detected by the listener's telephone 3d, and the voice heard by the listener is fed back to the speaker at a low volume with a changed voice quality, and the voice heard by the listener is recognized by the speaker. Because it does not disturb the speaker's utterance, it is possible to grasp the listener's response timing, eliminate the discomfort caused by the response delay, and adjust the speaker's utterance timing It is possible to eliminate discomfort during conversation.

<< Description of Fifth Embodiment >> FIG. 11 is a block diagram showing an example of an interactive communication system using an interactive speech speed conversion apparatus and method according to a fifth embodiment of the present invention. In this figure,
The same parts as those in FIG. 5 are denoted by the same reference numerals.

The difference between the interactive communication system 1e shown in this figure and the interactive communication system 1b shown in FIG. 5 is that the configuration of the telephone set 3b for transmitting the voice heard by the listener when the voice of the speaker is interrupted. Is provided with a telephone 3e partially modified.
According to e, when the listener is listening with a slow voice, the voice of the speaker is analyzed, and every time a pause section is detected, a voice message or signal notifying that the listener is listening with a slow voice is output by the speaker. This is to make it easier for the speaker to talk to the listener.

In this case, the telephone 3e on the listener side is
As shown in FIG. 2, while analyzing the voice from the speaker side for each short time block by the analysis processing unit 7, the characteristics of the voice are analyzed, and based on the analysis information output from the analysis processing unit 7, the speech speed conversion and synthesis are performed. The processing unit 8 converts the voice of the speaker to the speech speed, and allows the listener to listen to the slow voice. When the listener is listening to the slow voice, the signal generator 16 causes the listener to slow down. Causes a voice message or signal to be generated indicating that the user is listening.

In parallel with this operation, based on the analysis information output from the analysis processing unit 7, the silence determination unit 10 determines for each short-time block whether or not the speaker's voice is a silence block. When several blocks in the past are determined to be silent blocks by the silent determining unit 10 and the current block is determined to be a silent block, the pause determining unit 11 determines that the current block includes a pause, Based on the determination result, the signal switching unit 13
, A voice message or signal output from the signal generator 16 is selected and fed back to the speaker to make the speaker recognize whether or not the listener is listening to slow voice.

Thus, when the listener is currently listening to the slowly converted voice, and at the time when the utterance of the speaker is interrupted, the listener is currently listening to the slowly converted voice. The speaker is made to recognize and the next utterance timing is measured based on the recognition result, so that the talk between the speaker and the listener can be smoothly performed.

When the silence judging section 10 does not judge several past blocks as silence blocks, or when the current block is not judged as silence blocks,
The pause determination unit 11 determines that the current block does not include a pause. Based on the determination result, the signal switching unit 13 selects the zero signal output from the zero signal transmission unit 12 at all times. In this case, silence is fed back to the speaker so as not to disturb the utterance of the speaker.

As described above, in the fifth embodiment,
When the speaker's utterance continues, the speaker's telephone 2 is prohibited from transmitting a voice message or signal indicating whether or not the listener is listening to the slow sound, so that the speaker's utterance is not interrupted. When the utterance of the speaker is interrupted, this is detected by the telephone 3e on the listener's side, and when the listener is listening to the slow voice, a sound indicating this is transmitted to the speaker's side. Side, so that the listener can recognize that the listener is listening to the slow sound, let the listener understand the response timing of the listener while not disturbing the utterance of the speaker, and feel the discomfort caused by the response delay. This can be eliminated, and the speaker can adjust the utterance timing to eliminate discomfort during conversation.

<< Description of Sixth Embodiment >> FIG. 13 is a block diagram showing an example of an interactive communication system using an interactive speech speed conversion apparatus and method according to a sixth embodiment of the present invention. In this figure,
The same parts as those in FIG. 5 are denoted by the same reference numerals.

The difference between the interactive communication system 1f shown in this figure and the interactive communication system 1b shown in FIG. 5 is that the configuration of the telephone set 3b for transmitting the voice heard by the listener when the voice of the speaker is interrupted. Is provided with a partially modified telephone 3f.
f, the voice of the speaker is analyzed, and when a pause section is detected, the voice heard by the listener (voice obtained by converting the voice generated by the speaker into speech speed) and the voice of the listener are synthesized. Feedback to the speaker, the voice heard by the listener, the voice spoken by the listener,
For example, the voice of "Please speak more slowly" is recognized, and the call between the speaker and the listener is made smoother.

In this case, the telephone 3f on the listener side is
As shown in FIG. 4, while analyzing the voice from the speaker side for each short time block by the analysis processing unit 7, the characteristics of the voice are analyzed, and based on the analysis information output from the analysis processing unit 7, the speech speed conversion synthesis is performed. The processing section 8 converts the voice of the speaker into a speech speed, and allows the listener to hear a slow voice, and the volume control section 9 lowers the volume of the voice output from the speech speed conversion / synthesis processing section 8.

In parallel with this operation, based on the analysis information output from the analysis processing unit 7, the silence determination unit 10 determines for each short-time block whether or not the speaker's voice is a silence block. When several blocks in the past are determined to be silent blocks by the silent determining unit 10 and the current block is determined to be a silent block, the pause determining unit 11 determines that the current block includes a pause, Based on the determination result, the signal switching unit 13
Selects the sound of which the sound volume has been adjusted (the sound that the listener has listened to and the speech speed of which has been converted) output from the sound volume adjuster 9, and the signal adder 17 compares this sound with the sound that the listener has uttered. The speech is synthesized and fed back to the speaker, so that the speaker recognizes the voice heard by the listener and the voice uttered by the listener.

Thus, at the time when the utterance of the speaker is interrupted, the listener is made to recognize which part of the slowly converted voice is currently being heard, and at this time, what kind of request is made by the listener To recognize
Based on these recognition results, the next utterance timing is measured, and the call between the speaker and the listener is smoothly performed.

When the silence judging section 10 judges that some past blocks have not been judged as silence blocks, or when the current block has not been judged as silence blocks,
The pause determination unit 11 determines that the current block does not include a pause. Based on the determination result, the signal switching unit 13 selects the zero signal output from the zero signal transmission unit 12 at all times. In this case, silence is fed back to the speaker so as not to disturb the utterance of the speaker.

As described above, in the sixth embodiment,
When the speaker's utterance continues, the speaker's telephone 2 is prohibited from feeding back the voice heard by the listener, so that the speaker's utterance is not interrupted, and the speaker's utterance is interrupted. At this point, this is detected by the listener's telephone 3f, and the slower voice heard by the listener and the voice uttered by the listener are fed back to the speaker, and the voice heard by the listener is heard by the speaker. And recognize the voice uttered by the listener, so that the listener's response timing can be grasped without disturbing the speaker's utterance, and discomfort caused by the response delay can be eliminated. , Let the speaker adjust the utterance timing,
Discomfort during conversation can be eliminated.

<< Description of Another Embodiment >> In the above-described sixth embodiment, when the talk on the speaker side is interrupted, the sound output from the volume control section 9 by the signal switching section 13 (the sound that is heard by the listener side). 10), and synthesizes this voice with the voice of the listener, and transmits the synthesized voice to the speaker. The output side of the signal switching unit 13 shown in FIG. 8 and the signal shown in FIG. The output of the switching unit 13 or the output of the signal switching unit 13 shown in FIG.
Provided by the signal adding unit 17 and each signal switching unit 1
Alternatively, the voice (or voice message or signal) output from 3 and the voice of the listener may be synthesized and transmitted to the speaker.

Even in this case, as in the case of the sixth embodiment, when a conversation is performed between the speaker and the listener using the speech speed conversion function, the speaker finishes uttering. While waiting for the response of the listener, the speaker is notified of the voice indicating whether the listener is listening to the voice with the converted voice and the voice uttered by the listener, thereby preventing the speaker from uttering. In this way, it is possible to make the listener understand the response timing on the listener side and eliminate the discomfort caused by the response delay, and to adjust the utterance timing on the speaker side to eliminate the discomfort during conversation.

In each of the embodiments described above, the interactive speech speed conversion apparatus and method according to the present invention are described by taking the interactive communication systems 1a to 1f using a plurality of telephones 2, 3a to 3f as an example. The speech rate conversion apparatus and method according to the present invention may be applied to a speech rate conversion communicator, a speech rate conversion hearing aid, and the like.

Even in this manner, the conversation between the speaker and the listener can be smoothly performed without impairing the real-time property of the interactive speech speed conversion apparatus and method according to the present invention.

[0080]

[0081]

As described above, according to the present invention, in the interactive speech speed conversion device according to the first aspect, a conversation is performed between the speaker and the listener using the speech speed conversion function. When the speaker has finished speaking and is waiting for the response from the listener, the speaker informs the speaker of the speed-converted voice that the listener is listening to, so as not to disturb the speaker's utterance. In addition, it is possible to make the listener understand the response timing on the side of the listener, eliminate the sense of incongruity caused by the response delay, and adjust the utterance timing on the speaker side, thereby eliminating the feeling of incongruity during conversation.

According to the second aspect of the present invention, when the conversation is performed between the speaker and the listener using the speech speed conversion function, the speaker finishes uttering and the listener changes the speech. While waiting for a response, the listener extracts the low-frequency component of the speech-rate-converted voice that the listener is listening to and informs the speaker, thereby preventing the speaker's utterance while preventing the speaker's utterance. Not only can you understand the timing, you can eliminate the discomfort caused by the response delay,
The speaker can adjust the utterance timing to eliminate discomfort during conversation.

According to the third aspect of the present invention, when a conversation is performed between the speaker and the listener using the speech speed conversion function, the speaker finishes speaking and the listener changes the speech. While waiting for a response, change the characteristics of the speed-converted speech that the listener is listening to, let the speaker know, and do not disturb the speaker's utterance,
It is possible to make the listener understand the response timing on the listener side and eliminate the sense of incongruity caused by the response delay, and to adjust the utterance timing on the speaker side to eliminate the incongruity in conversation.

According to the fourth aspect of the present invention, when a conversation is performed between the speaker and the listener by using the speech speed conversion function, the speaker finishes uttering and the listener's side stops. While waiting for a response, the listener is informed that the listener is listening to the voice with the converted speech rate, so that the listener's response timing is grasped while not disturbing the speaker's utterance, and responding. The discomfort caused by the delay can be eliminated, and the speaker can adjust the utterance timing to eliminate the discomfort during conversation.

In the conversation type speech rate converter according to the fifth aspect, when a conversation between the speaker side and the listener side is performed by using the speech rate conversion function, the speaker side finishes uttering and the listener side finishes speaking. While waiting for a response, the speaker is informed of the sound or signal indicating whether the listener is listening to the speed-converted sound and the sound uttered by the listener so that the speaker's utterance is not interrupted. In addition, it is possible to make the listener understand the response timing on the listener side, to eliminate the sense of incongruity caused by the response delay, and to adjust the utterance timing on the speaker side to eliminate the incongruity during conversation.

In the interactive speech rate converter according to the sixth aspect, conversation is performed between the talker side and the listener side by using any one of a speech rate conversion telephone, a speech rate conversion communication device, and a speech rate conversion hearing aid. When the speaker has finished speaking and is waiting for a response from the listener, a voice or signal indicating whether the listener is listening to the speed-converted voice and, if necessary, the listener speaking Informs the speaker of the voice, thereby preventing the speaker's utterance from being disturbed, grasping the listener's response timing, eliminating discomfort caused by the response delay, and adjusting the speaker's utterance timing By doing so, it is possible to eliminate a sense of discomfort during conversation.

[0087]

In the interactive speech speed conversion method according to the present invention, when a conversation is performed between the speaker and the listener using the speech speed conversion function, the speaker finishes uttering, and the listener's speech ends. While waiting for a response, the listener informs the speaker of the speed-converted voice that the listener is listening to, so as not to disturb the speaker's utterance, let the listener understand the response timing, and respond to the delay. The discomfort caused by the problem can be eliminated, and the speaker can adjust the utterance timing to eliminate the discomfort during the conversation.

In the conversation type speech rate conversion method according to the present invention, when a conversation between the speaker side and the listener side is performed using the speech rate conversion function, the speaker side finishes uttering, and the listener side finishes speaking. While waiting for a response, the listener extracts the low-frequency component of the speech-rate-converted voice that the listener is listening to and informs the speaker, thereby preventing the speaker's utterance while preventing the speaker's utterance. Not only can you understand the timing, you can eliminate the discomfort caused by the response delay,
The speaker can adjust the utterance timing to eliminate discomfort during conversation.

In the interactive speech speed conversion method according to the ninth aspect, when a conversation is performed between the speaker and the listener using the speech speed conversion function, the speaker finishes speaking and the listener's While waiting for a response, change the characteristics of the speed-converted speech that the listener is listening to, let the speaker know, and do not disturb the speaker's utterance,
It is possible to make the listener understand the response timing on the listener side and eliminate the sense of incongruity caused by the response delay, and to adjust the utterance timing on the speaker side to eliminate the incongruity in conversation.

According to the tenth aspect of the present invention, when a conversation is performed between the speaker and the listener using the speech speed conversion function, the speaker finishes uttering and the listener's side. While waiting for a response, the listener is informed that the listener is listening to the voice with the converted speech rate, so that the listener's response timing is grasped while not disturbing the speaker's utterance, and responding. The discomfort caused by the delay can be eliminated, and the speaker can adjust the utterance timing to eliminate the discomfort during conversation.

In the interactive speech speed conversion method according to the eleventh aspect, when a conversation is being performed between the speaker and the listener using the speech speed conversion function, the speaker finishes uttering and the listener ends. While waiting for a response, the speaker is informed of the sound or signal indicating whether the listener is listening to the speed-converted sound and the sound uttered by the listener so that the speaker's utterance is not interrupted. In addition, it is possible to make the listener understand the response timing on the listener side, to eliminate the sense of incongruity caused by the response delay, and to adjust the utterance timing on the speaker side to eliminate the incongruity during conversation.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an example of an interactive communication system using a first embodiment of an interactive speech speed conversion device and method according to the present invention.

FIG. 2 is a block diagram showing a detailed configuration example of a telephone on the listener side shown in FIG. 1;

FIG. 3 is an explanatory diagram schematically showing an operation example of the telephone on the listener side shown in FIG. 2;

FIG. 4 is a block diagram showing an example of an interactive communication system using a second embodiment of the interactive speech speed conversion device and method according to the present invention.

FIG. 5 is a block diagram showing a detailed configuration example of a telephone on the listener side shown in FIG. 4;

6 is a timing chart showing an operation example of the interactive communication system shown in FIG.

FIG. 7 is a block diagram showing an example of an interactive communication system using a third embodiment of the interactive speech speed conversion device and method according to the present invention.

8 is a block diagram showing a detailed configuration example of the telephone set on the listener side shown in FIG. 7;

FIG. 9 is a block diagram showing an example of an interactive communication system using a fourth embodiment of the interactive speech speed conversion device and method according to the present invention.

FIG. 10 is a block diagram showing a detailed configuration example of a telephone on the listener side shown in FIG. 9;

FIG. 11 is a block diagram showing an example of an interactive communication system using a fifth embodiment of the interactive speech speed conversion device and method according to the present invention.

FIG. 12 is a block diagram showing a detailed configuration example of a telephone on the listener side shown in FIG. 11;

FIG. 13 is a block diagram illustrating an example of an interactive communication system using a sixth embodiment of the interactive speech speed conversion device and method according to the present invention.

14 is a block diagram showing a detailed configuration example of the telephone on the listener side shown in FIG.

[Explanation of symbols]

1a, 1b, 1c, 1d, 1e, 1f: Interactive communication system 2, 3a, 3b, 3c, 3d, 3e, 3f: Telephone 4: Line 5: Reception detection unit (conversation start detection unit) 6: Message signal transmission Unit (notification unit) 7: analysis processing unit (conversation interruption detection unit) 8: speech speed conversion / synthesis processing unit 9: volume adjustment unit (notification unit) 10: silence determination unit (conversation interruption detection unit) 11: pause determination unit ( Conversation interruption detection unit) 12: Zero signal transmission unit 13: Signal switching unit (notification unit) 14: Low-pass filtering unit (notification unit) 15: Voice quality processing unit (notification unit) 16: Signal generation unit (notification unit) 17 : Signal addition unit (notification unit)

──────────────────────────────────────────────────続 き Continuation of front page (72) Inventor Tohru Toki 1-10-11 Kinuta, Setagaya-ku, Tokyo Japan Broadcasting Corporation Research Institute (56) References JP-A-6-311211 (JP, A) ( 58) Field surveyed (Int.Cl. ⁷ , DB name) H04M 1/00 G10L 19/00 G10L 21/04

Claims (11)

(57) [Claims] 1. An interactive speech speed conversion apparatus for converting a speech speed spoken by a speaker and listening to the speech, wherein the speech of the speaker is terminated when the conversation is performed using the speech speed conversion function, and A conversation interruption detection unit that detects a state of waiting for a response, and when the interruption of conversation is detected by the conversation interruption detection unit, the speech speed converted voice heard by the listener is directly reduced or the volume is reduced. And a notifying unit for notifying the speaker of the conversation. 2. An interactive speech rate conversion device for converting a speech rate spoken by a speaker and listening to the speech, wherein the speech on the speaker side is terminated and the listener A conversation interruption detection unit that detects a state of waiting for a response from the listener, and when the conversation interruption is detected by the conversation interruption detection unit, a low-frequency component of a speech-speed-converted voice that is heard by a listener is extracted. And a notifying unit for notifying the speaker at the same volume or at a reduced volume thereafter. 3. An interactive speech rate conversion apparatus for converting a speech rate spoken by a speaker and listening to the speech, wherein the speech of the speaker is terminated when the conversation is performed using the speech rate conversion function, and A conversation interruption detection unit for detecting a state of waiting for a response, and when the conversation interruption is detected by the conversation interruption detection unit, after changing the characteristics of the speech speed converted voice heard by the listener, A notification unit for notifying a speaker at the same volume or a reduced volume, and a speaker unit. 4. An interactive speech rate conversion device for converting a speech rate spoken by a speaker and listening to the speech, wherein the speech of the speaker is terminated when the conversation is performed using the speech rate conversion function, and A conversation interruption detecting unit for detecting a state of waiting for a response from the user, and a voice message indicating that the listener is listening to the speech speed converted sound when the conversation interruption is detected by the conversation interruption detecting unit. Or a notifying unit for notifying a speaker of a signal. 5. The interactive speech rate conversion device according to claim 1, wherein when the listener notifies the speaker that he / she is listening to the speech whose speech rate has been converted, the speech uttered by the listener. An interactive speech speed conversion device, comprising: 6. The conversation type speech speed conversion device according to claim 1, wherein the conversation interruption detection unit and the notification unit are a speech speed conversion telephone on the listener side and a speech speed conversion communication on the listener side. An interactive speech speed conversion device, which is provided in one of a telephone and a listener's speech speed conversion hearing aid. 7. An interactive speech speed conversion method in which a speech speed converted by a speaker is listened by converting the speech speed. In a conversation using a speech speed conversion function, the speaker's speech ends and the listener's speech is terminated. A conversational voice speed conversion method comprising: detecting a state of waiting for a response from the speaker; and notifying the voice of the voice speed converted by the listener as it is or by reducing the volume to the speaker side. 8. An interactive speech rate conversion method for converting a speech rate spoken by a speaker and listening to the speech, wherein the speech of the speaker is terminated when the conversation is performed using the speech rate conversion function, and Detecting the state of waiting for a response from the speaker, extracting the low-frequency component of the speech-rate-converted voice heard by the listener, and then notifying the speaker at the same volume or a reduced volume. Interactive speech rate conversion method. 9. An interactive speech rate conversion method in which a speech rate converted by a speaker is listened by converting the speech rate. In a conversation using a speech rate conversion function, the speaker's speech ends and the listener's speech is terminated. Detecting the state of waiting for a response from the listener, changing the characteristics of the speech speed converted voice heard by the listener, and then notifying the speaker at the same volume or a reduced volume. Interactive speech rate conversion method. 10. An interactive speech rate conversion method in which a speaker converts a speech rate spoken by a speaker and listens to the speech. In a conversation using a speech rate conversion function, the speaker's speech ends and the listener's speech is terminated. An audio message or signal indicating that the listener is listening to the speech speed-converted voice is notified to the speaker side, and a state of waiting for a response from the speaker is detected. 11. The interactive speech rate conversion method according to claim 7, wherein when notifying the listener that the listener is listening to the speech rate-converted speech, the speech uttered by the listener. Sending to the speaker's side.