JP2019087798A - Voice input device - Google Patents

Abstract

To appropriately transmit uttered voice even when a low-volume voice pronunciation is performed.SOLUTION: A drive control unit 130 generates a vibration given to the user's throat by an actuator 131. In response to this vibration, a low-volume voice can be pronounced from the user's mouth such that the surrounding people cannot hear the voice. A microphone 140 picks up this voice and outputs a voice signal. A voice conversion unit 150 performs voice conversion processing to facilitate recognition of the voice signal. Specifically, the voice conversion unit 150 performs voice recognition processing of the voice signal to output character information, further synthesizes a voice signal from the character information, or replace the waveform of a consonant section of the voice signal with a voice signal waveform of the appropriate consonant obtained by the voice recognition processing.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a voice input device suitable for voice input in a cellular phone or the like.

  There is an electric artificial larynx as a technical means for making a person without a vocal cord pronounce a voice (see, for example, Patent Document 1). The electric artificial larynx is a device that vibrates a person's throat and produces an air vibration similar to that caused by the vibration of the vocal cords in the oral cavity. The user can produce sound from the mouth by giving vibration to the throat by the electric artificial larynx and changing the shape of the mouth. A person without a vocal cord can use this electric artificial larynx to make a normal conversation and to make a call using a telephone.

International Publication No. 1999/12501

  By the way, in a telephone, an audio signal collected by a microphone is transmitted as it is. Therefore, when the collected voice signal is a difficult voice signal, the voice signal heard by the other party is also hard to hear. As an example of a situation where such a situation occurs, there is a case where a mobile phone makes a call with a low voice so as not to be heard by people around the train. Here, when the user of the mobile phone utters a low voice, the voice signal obtained by the microphone of the mobile phone has an extremely small level, and the voice signal waveform is an audio signal waveform at the time of tone generation at a normal volume. It will be distorted compared. Therefore, even if the voice signal obtained by the microphone is sent to the calling party, it is difficult for the calling party to recognize what the user of the mobile phone is talking. This is a problem that occurs both when a person without a vocal cord talks with a small voice using the electric artificial larynx and when a healthy person talks with a low voice.

  The present invention has been made in view of the above circumstances, and it is an object of the present invention to provide a technical means capable of appropriately transmitting the voiced voice even when a low voiced voice is produced. To aim.

  The present invention relates to a voice acquiring means for acquiring a voice signal indicating a voice to be produced from the mouth of the user according to a vibration given to the throat of the user, and a voice conversion process for facilitating recognition of the voice signal. A voice input device is provided which comprises: voice conversion means for applying.

  According to the present invention, even if the volume of the audio signal acquired by the audio acquisition means is low enough that the surrounding people can not hear it, the audio conversion means converts the audio signal into information for easy recognition. Be done. Therefore, even when a low voice is produced, the voiced voice can be properly transmitted.

It is a block diagram which shows the structure of the mobile telephone containing the audio | voice input device which is one Embodiment of this invention. It is a block diagram which shows the structure of the speech conversion part in the embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing the configuration of a mobile telephone 1000 including a voice input device 100 according to an embodiment of the present invention. FIG. 1 illustrates, in addition to the voice input device 100, the transmission unit 201 and the reception unit 202 of the mobile phone 1000, and the head 10 of the user of the mobile phone 1000. Although the mobile phone 1000 includes various devices similar to an ordinary mobile phone such as an antenna, the illustration thereof is omitted in FIG.

  In the voice input device 100, the sound insulation hood 132 and the actuator 131 accommodated therein are fixed to the neck of the user by the belt 133. Here, the actuator 131 functions as a means for bringing its vibrating surface into contact with the user's throat and applying vibration to the user's throat. The sound insulation hood 132 is a means for shielding so that the vibration noise of the actuator 131 does not leak to the surroundings.

  The tilt sensor 151 is an ear-hook type sensor, which is attached to the user's ear of the mobile phone 1000 and detects the tilt of the user's head.

  The sound generation start button 134 is a push button for instructing start of sound generation. The user of the portable telephone 1000 holds the operator provided with the sound generation start button 134 in hand, and turns on the sound generation start button 134 when sound generation is started. The sound generation start button 134 may be a button switch that is turned on only during a pressed period, or may be a button switch that is turned from off to on and from on to off each time it is pressed.

  The control unit 110 functions as a control center of the voice input device 100 and also functions as a control center of the entire mobile phone 1000.

  The operation display unit 120 is, for example, a touch panel, and is a device that displays various types of information to the user and receives an operation of the user.

  The drive control unit 130 is a drive control unit that applies a periodic drive pulse waveform to the actuator 131 under the control of the control unit 110 and thereby generates a vibration to be applied to the user's throat. More specifically, the control unit 110 causes the drive control unit 130 to start outputting the drive pulse waveform when the sound generation start button 134 is turned on.

  While vibration is applied to the throat from the actuator 131, the user can produce a desired voice by opening and closing the mouth while changing the shape of the oral cavity 1. The microphone 140 is a means for collecting the user's voice as in the case provided in a normal cellular phone, but in the present embodiment, in addition to this, according to the vibration given to the user's throat It functions as an audio acquisition means for acquiring an audio signal indicating an audio generated from the user's mouth.

  In the present embodiment, it is possible to control the volume of sound generation in sound generation using the actuator 131. More specifically, in the present embodiment, the user can set the pulse width of the drive pulse waveform provided to the actuator 131 by the drive control unit 130 by operating the operation display unit 120. Here, when the pulse width of the drive pulse waveform is increased, the volume of the sound produced from the user's mouth is increased. Further, if the pulse width of the drive pulse waveform is shortened, the volume of the sound produced from the user's mouth will be reduced. When the pulse width of the drive pulse waveform is shorter than a predetermined length, the volume of the sound produced from the user's mouth can be set to a small volume that can not be heard by the surrounding people. In general, when the electric artificial larynx is used for conversation, the electric artificial larynx is driven with large amplitude in order to secure the volume of the voiced speech. However, when making a call by the mobile phone 1000 in a train, the actuator 131 is driven with a small amplitude to make the volume of the sound produced from the user's mouth small enough to be heard by the surrounding people.

  Further, in the voice input device 100 according to the present embodiment, the intonation according to the change in the inclination of the head of the user, that is, the voice generated from the user's mouth making a sound using the vibration of the actuator 131 , Has a function to give a change in pitch.

  More specifically, the output signal of the inclination sensor 151 is supplied to the control unit 110. The control unit 110 generates intonation information indicating the pitch conversion ratio of the voice produced by the user based on the output signal of the tilt sensor 151. The pitch conversion ratio is a ratio of the pitch of the voice after pitch conversion to the pitch before pitch conversion.

  Here, the inclination angle in a state where the front of the head of the user is directed to the front is 0 °. When the head front surface faces upward, the inclination angle changes from 0 ° to the positive direction, and when the head front surface faces downward, the inclination angle changes from 0 ° to the negative direction.

  In this case, when the inclination angle changes from 0 ° to the positive direction, the control unit 110 increases the pitch conversion ratio indicated by the intonation information from 1 and when the inclination angle changes from 0 ° to the negative direction, the control unit 110 changes the intonation information. Decreases the pitch conversion ratio shown by.

  The drive control unit 130 controls the cycle of the drive pulse waveform given to the actuator 131 based on the intonation information output by the control unit 110. When the pitch conversion ratio indicated by the intonation information is 1, the drive control unit 130 provides the actuator 131 with a drive pulse waveform of a standard cycle. As a result, a standard pitch voice is produced from the user's mouth. When the pitch conversion ratio indicated by the intonation information increases from 1, the drive control unit 130 shortens the cycle of the drive pulse waveform given to the actuator 131 according to the increase in the pitch conversion ratio. As a result, the pitch of the sound produced from the user's mouth rises from the standard pitch. Also, when the pitch conversion ratio indicated by the intonation information decreases from 1, the drive control unit 130 lengthens the cycle of the drive pulse waveform given to the actuator 131 according to the decrease in the pitch conversion ratio. As a result, the pitch of the sound produced from the user's mouth drops from the standard pitch.

  The voice conversion unit 150 is a means for performing voice conversion processing to facilitate recognition of the voice signal output from the microphone 140. As this voice conversion process, three modes of voice / character conversion mode, voice / voice conversion mode and voice processing mode are prepared. Here, the voice / character conversion mode is a mode in which voice recognition processing is performed on the voice signal output from the microphone 140 and character information is output. The voice / voice conversion mode is a mode in which voice recognition processing is performed on the voice signal output from the microphone 140, and the voice signal is synthesized based on the character information obtained as a result. In the voice processing mode, voice recognition processing is performed on the voice signal output from the microphone 140, and the voice signal output from the microphone 140 is processed into a voice signal easy to recognize based on the result of the voice recognition processing. It is a mode. In the present embodiment, the user of the mobile phone 1000 can select a desired mode from the three types of modes by operating the operation display unit 120 and can cause the voice conversion unit 150 to execute the desired mode.

  FIG. 2 is a block diagram showing the configuration of the speech conversion unit 150. As shown in FIG. As shown in FIG. 2, the speech conversion unit 150 includes a speech recognition unit 151, a speech synthesis unit 152, a speech processing unit 153, and a switch 154.

  The voice recognition unit 151 performs voice recognition of the voice signal output from the microphone 140 and outputs character information. Here, when the pulse width of the drive pulse waveform given to the actuator 131 is set short and the sound produced from the user's mouth is small enough to be heard by the surrounding people, the sound output from the microphone 140 The level of the vowel section of the signal is very small and the level of the consonant section is even smaller. In this case, although recognition of vowels in the speech signal is possible, recognition of consonants is difficult.

  Therefore, the speech recognition unit 151 in the present embodiment performs speech recognition by determining a word break in the speech signal based on intonation (pitch change) of the speech signal. Specifically, the speech recognition unit 151 stores a database of pitch change patterns of speech measured for various words used for conversation. Then, when a pitch change pattern that matches any of the pitch change patterns in the database appears in the sound signal output from the microphone 140, the voice recognition unit 151 determines the section of the sound signal corresponding to the pitch change pattern. Treat as one word and execute speech recognition.

  Further, the speech recognition unit 151 in the present embodiment refers to the speech recognition result obtained up to the present time while executing the speech recognition process, and the section in which the recognition is not completed (a section in which the unknown phoneme of the type exists) Perform voice recognition). For example, in speech recognition processing, it is assumed that the type of consonant in a section where a speech signal is present is unknown. In this case, in the speech recognition process, the type of the consonant is estimated from the context indicated by the character that is the speech recognition process result of the sections before and after the section. Specifically, several consonant candidates are selected based on the speech signal of the section where the type of consonant is unknown. Then, one consonant is selected from the consonant candidates one by one and applied to the section, and it is determined whether characters in a section including the section and the sections before and after the section constitute a meaningful sentence. Then, consonants that constitute meaningful sentences are selected as speech recognition results.

  The voice synthesis unit 152 is means for synthesizing a voice signal based on the text information output from the voice recognition unit 151. Specifically, the speech synthesis unit 152 stores a database of speech waveforms of various speech segments such as consonants and vowels, and reads speech waveforms of speech segments of consonants and vowels indicated by character information from the database. The voice signal is synthesized by connecting on the time axis.

  In a preferred embodiment, a speech synthesis unit 152 stores a database of speech waveforms of speech segments corresponding to various kinds of speech such as husky male speech and highly transparent female speech. The user can select a desired type of database by the operation of the operation display unit 120 and use it for speech synthesis.

  The speech synthesis unit 152 in the present embodiment has means for controlling the pitch of the synthesized speech signal based on intonation information. Here, the intonation information is information generated by the control unit 110 based on the output signal of the inclination sensor 151, and indicates a change in the pitch of the sound produced from the user's mouth. Therefore, the voice signal output from the voice synthesis unit 152 is a voice signal having a pitch change similar to the pitch change of the voice produced from the user's mouth.

  The voice processing unit 153 is a means for processing the voice signal output from the microphone 140 into a voice signal that can be easily recognized based on the voice recognition processing result of the voice recognition unit 151. As described above, when the user uses the actuator 131 to produce sounds with a small volume that can not be heard by the surrounding people, the volume of the sound signal output from the microphone 140, especially in the consonant section, is extremely small and difficult to recognize. is there. Therefore, in the present embodiment, the speech processing unit 153 stores a database of speech signal waveforms representing various consonants in advance. Then, in the audio signal output from the microphone 140, the audio processing unit 153 selects an audio signal waveform of a section determined as a consonant by the speech recognition processing of the speech recognition unit 151 as an appropriate audio signal corresponding to the consonant in the database. Replace with waveform. Further, the audio processing unit 153 amplifies and outputs the audio signal after replacement of the audio signal waveform of the consonant to an appropriate level for easy listening.

  The switch 154 selects one of the character information output from the speech recognition unit 151, the speech signal output from the speech synthesis unit 152, and the speech signal output from the speech processing unit 153, and outputs the selected signal to the transmission unit 201 shown in FIG. It is a means. When the speech conversion unit 150 is set to the speech / character conversion mode, the speech recognition unit 151 is activated, and the character information output from the speech recognition unit 151 is supplied to the transmission unit 201 via the switch 154. When the voice conversion unit 150 is set to the voice / voice conversion mode, the voice recognition unit 151 and the voice synthesis unit 152 are activated, and the voice signal output from the voice synthesis unit 152 is supplied to the transmission unit 201 via the switch 154. Be done. When the speech conversion unit 150 is set to the speech processing mode, the speech recognition unit 151 and the speech processing unit 153 are activated, and the speech signal output from the speech processing unit 153 is supplied to the transmission unit 201 via the switch 154. .

  In FIG. 1, the transmission unit 201 is a means for transmitting an audio signal or text information to a calling party. When the voice conversion unit 150 is set to the voice / character conversion mode, the transmission unit 201 transmits the character information output from the voice conversion unit 150 to the telephone set of the other party of the user of the mobile telephone 1000. When the voice conversion unit 150 is set to the voice / voice conversion mode or the voice processing mode, the transmission unit 201 transmits the voice signal output from the voice conversion unit 150 to the telephone of the other party of the user of the mobile telephone 1000. Do. The portable telephone 1000 negotiates with the other party's telephone when starting a call with the other party's telephone. At this time, the cellular phone 1000 transmits information indicating whether the transmission unit 201 transmits character information or voice signal to the telephone of the other party. As a result, the telephone of the other party of the call can display when the character information is sent, and can emit the sound when the voice signal is sent.

  The receiving unit 202 is means for receiving an audio signal from the telephone of the other party. The audio signal received by the receiving unit 202 is sent to the speaker 170 via the adder 163 and emitted by the speaker 170.

In the present embodiment, the user of the mobile phone 1000 can monitor the processing result of the voice conversion unit 150 by operating the operation display unit 120. For example, in the state where the voice / character conversion mode is set, the user can turn on the switch 162 by operating the operation display unit 120. As a result, character information output from the voice recognition unit 151 of the voice conversion unit 150 is sent to the operation display unit 120 via the switch 162 and displayed on the operation display unit 120. Further, in a state where the voice / voice conversion mode or the voice processing mode is set, the user can turn on the switch 161 by operating the operation display unit 120. As a result, the audio signal output from the audio synthesis unit 152 or the audio processing unit 153 of the audio conversion unit 150 is sent to the speaker 170 via the switch 161 and the adder 163 and emitted from the speaker 170.
The above is the details of the mobile phone 1000 according to the present embodiment.

  In the present embodiment, the user of the mobile phone 1000 sets the pulse width of the drive pulse waveform given to the actuator 131 to the minimum value by the operation of the operation display unit 120, for example, when making a call by the mobile phone 1000 in a train. . Then, the user turns on the sound generation start button 134 to change the shape of the oral cavity 1. As a result, the voice is produced at a volume that can not be heard by the surrounding person from the user's mouth, and a voice signal indicating this voice is output from the microphone 140. At this time, the user can change intonation information by tilting the head, and can give intonation, that is, change of the pitch to the sound pronounced from the user's mouth.

  When the voice / character conversion mode is set, the voice recognition unit 151 of the voice conversion unit 150 performs voice recognition of the voice signal output from the microphone 140, and outputs character information. Here, the sound signal output from the microphone 140 has a low level, and in particular, it is difficult to recognize consonants. Therefore, the speech recognition unit 151 determines the break of words in the speech signal based on the pitch change appearing in the speech signal, and estimates the consonant based on the context indicated by the characters before and after the consonant in the speech recognition process. Generates character information from the speech signal. The character information output from the voice recognition unit 151 is sent to the telephone of the other party by the transmission unit 201, and displayed by the telephone.

  When the voice / voice conversion mode is set, the voice recognition unit 151 of the voice conversion unit 150 performs voice recognition processing of the voice signal output from the microphone 140, and outputs character information. Then, the speech synthesis unit 152 synthesizes the speech signal from the character information, and controls and outputs the pitch of the speech signal based on the intonation information. Then, the voice signal output from the voice synthesis unit 152 is sent to the telephone of the other party by the transmission unit 201, and the speaker of the telephone emits the sound.

  When the voice processing mode is set, the voice recognition unit 151 of the voice conversion unit 150 performs voice recognition processing of the voice signal output from the microphone 140. Then, based on the speech recognition processing result, the speech processing unit 153 replaces the speech signal waveform of the consonant period in the speech signal output from the microphone 140 with the speech signal waveform of the appropriate consonant, and the speech after this substitution The signal is amplified to an appropriate level and output. Then, the voice signal output from the voice processing unit 153 is sent by the transmission unit 201 to the telephone of the other party, and the speaker of the telephone emits the sound.

  As described above, according to the present embodiment, the user of the mobile phone 1000 uses the vibration of the actuator 131 to perform sound production at a small volume that can not be heard by the surrounding people, and the voice of the mobile phone 1000 is output. Audio can be input to the input device 100. Then, even when the input of the voice with a small voice is performed as described above, in the voice input device 100, the voice conversion unit 150 performs a voice conversion process for facilitating the recognition on the voice signal obtained by the microphone 140 and transmits it. The unit 201 transmits to the other party. Therefore, the user can appropriately transmit desired information to the other party even if the user utters a low voice.

  Further, in the present embodiment, in the voice / character conversion mode, the voice signal obtained by the microphone 140 can be converted into character information and transmitted to the other party. Therefore, the user can accurately transmit desired information to the other party.

  Further, in the present embodiment, in the voice / voice conversion mode, the voice signal obtained by the microphone 140 can be converted into character information, and the voice signal can be synthesized from the character information and transmitted to the other party. Therefore, the user can transmit desired information to the other party in a form close to that of a normal call.

  Further, in the present embodiment, in the sound processing mode, only the sound signal waveform of the consonant section of the sound signal obtained by the microphone 140 can be replaced with the sound signal waveform of appropriate consonant and transmitted to the other party. Therefore, the user can transmit the voice close to the voice pronounced by the user to the other party. It is suitable for users who want to deliver their voice to the other party.

  In addition, the voice input device 100 according to the present embodiment can also be used by healthy persons with vocal cords. Therefore, both the person without the vocal cords and the normal subject with the vocal cords use the cellular phone 1000 equipped with the voice input device 100 to make a low-voice call that can not be heard by the surrounding people. Since a person without a vocal cord talks in the same manner as a normal person with a vocal cord in this way, the person without the vocal cord can be motivated to talk by the mobile phone 1000.

Other Embodiments
As mentioned above, although each embodiment of this invention was described, other embodiments can be considered to this invention. For example:

(1) In the above embodiment, the period of the drive pulse waveform given to the actuator 131 is controlled by the intonation information obtained from the output signal of the inclination sensor 151, but it is not necessary to control this period. In this case, the speech recognition unit 151 may determine the break of words in the speech signal output from the microphone 140 based on the pitch change indicated by the intonation information.

(2) In the above embodiment, intonation information is generated by the inclination sensor 151 mounted on the head of the user. However, intonation information may be generated by a sensor other than the tilt sensor. For example, an acceleration sensor or the like may be attached to a body part of the user, and the intonation information may be generated by this sensor. In addition, the sensor for generating intonation information may be attached to a part other than the head of the user. In addition, the user may operate an operator such as a slide switch to generate intonation information.

(3) In the voice / character conversion mode of the above embodiment, the intonation information may be transmitted to the telephone of the other party in synchronization with the character information output from the voice recognition unit 151. In this case, the telephone of the other party of the call synthesizes a voice signal from the received character information, and controls the pitch of the voice signal obtained from this character information based on intonation information received in synchronization with the character information. May be

(4) In the voice / voice conversion mode of the above embodiment, instead of using the intonation information obtained from the output signal of the inclination sensor 151, the pitch ratio of the pitch of the voice signal output from the microphone 140 to the standard pitch Intonation information may be generated, and the pitch of the speech signal synthesized by the speech synthesis unit 152 may be controlled based on the intonation information.

(5) An operator may be provided for instructing the timing of sound generation of the beginning of the word, and the speech recognition unit 151 may detect the timing of the beginning of the speech signal based on the operation of the operator and perform speech recognition. Alternatively, the sound generation start button 134 in the above embodiment may be used as an operator for instructing the sound generation timing of this word head.

(6) In the voice / voice conversion mode, a voice signal is synthesized from character information obtained by voice recognition processing, but as a result obtained in the process of voice recognition processing, for example, voice signals may be synthesized from formant information etc. .

(7) In the speech processing mode, the speech signal may be processed by simple waveform conversion processing not using speech recognition processing. For example, a consonant interval in the audio signal is detected based on the level and frequency of the audio signal obtained from the microphone 140, and level conversion processing and consonant enhancement processing are performed only for the detected consonant interval. By performing such audio processing, the audio signal can be brought close to the audio signal of normal audio.

(8) The voice input device 100 transmits the voice signal obtained from the microphone 140 to the server via the network, and causes the server to execute the processing of the voice conversion unit 150, and the voice signal and character information as the execution result from the server It may be received and transmitted to the other party by the transmitting unit 201. According to this aspect, since it is not necessary to provide the voice conversion unit 150 in the mobile phone 1000, an increase in cost of the mobile phone 1000 can be avoided. In order to achieve the object of the present invention, there is provided an audio acquisition process for acquiring an audio signal indicating an audio generated from the user's mouth according to a vibration given to the user's throat, and recognition of the audio signal. Any device may perform speech conversion processing to facilitate. The device for executing the voice acquisition process and the device for executing the voice conversion process may be any devices and arranged.

(9) Although the voice input device according to the present invention is used for a telephone in the above embodiment, the application of the voice input device according to the present invention is not limited to this. The voice input device according to the present invention is applicable to all voice input devices including, for example, a text input device using voice.

Reference numeral 1000: mobile phone, 201: transmission unit, 202: reception unit, 100: voice input device, 110: control unit, 120: operation display unit, 130: drive control unit, 131: actuator, 132: sound insulation hood, 133: belt, 140: microphone, 150: voice conversion unit, 161, 163, 154: switch, 163: adder, 151: voice recognition unit, 152: voice synthesis Part, 153: voice processing part, 10: head, 1: oral cavity.

Claims (7)

Voice acquisition means for acquiring a voice signal indicating a voice generated from the user's mouth according to a vibration given to the user's throat;
A voice conversion unit for performing voice conversion processing to facilitate recognition of the voice signal.   2. The voice input device according to claim 1, further comprising drive control means for generating vibration to be applied to the throat of the user.   3. The voice input device according to claim 1, wherein the voice conversion unit performs voice recognition processing of the voice signal to generate character information.   3. The voice input device according to claim 1, wherein the voice conversion unit performs voice recognition processing of the voice signal, and synthesizes a voice signal based on a voice recognition processing result.   The voice according to claim 1 or 2, wherein the voice conversion means performs voice recognition processing of the voice signal, and processes the voice signal into a voice signal that is easy to recognize based on the voice recognition processing result. Input device. The voice input means comprises an operator for operating the cycle of the vibration;
The voice conversion unit according to any one of claims 3 to 5, wherein the voice conversion unit divides the voice signal on a time axis based on a change in pitch of the voice signal and performs the voice recognition process. Voice input device as described.   The speech according to any one of claims 3 to 6, wherein the speech conversion means executes the speech recognition process based on the result of the character information obtained by the speech recognition process. Input device.

Images