JP2002073605A - Method for interpreting telephone voice - Google Patents

Abstract

PROBLEM TO BE SOLVED: To facilitate communication by an international telephone call between speakers having different linguistic cultures from each other. SOLUTION: Speech communication equipment 1 is provided with a voice recognizing device 3 for identifying a voice inputted from a telephone line 10 and outputting character data decided in accordance with recognition data, a translating device 2 for converting the character data into a language decided in accordance with the designated opposite speech party and outputting the language as language data, and a voice synthesizing device 4 for outputting a voice signal combined in accordance with the inputted language data, and performs interpretation between the speakers connected to the telephone line and having different languages.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a telephone voice interpreting method, and more particularly to a telephone voice interpreting method for holding a conversation with a specified foreign language partner accommodated in a telephone network in a native language.

[0002]

2. Description of the Related Art With the recent globalization, opportunities for making international calls have increased. The biggest barrier to making international calls is the language barrier. If you are not familiar with the foreign language of the other party, you will not be able to communicate easily.

[0003] An automatic interpreting telephone system capable of transmitting and receiving a conversation in an international telephone in its own language and receiving in its own language is a speech data recognized as a speech recognition device for telephone speech by an interpreting device accommodated in a telephone network. And a speech synthesizer for synthesizing the translated foreign language into a voice signal.

[0004]

It is essential for the interpreter of the above-described conventional telephone voice automatic interpreter telephone system to solve problems such as updating of speech recognition information and speech synthesis information accompanying the addition of a corresponding language. is there.

[0005] It is an object of the present invention to provide a telephone voice interpreting method that enables smooth updating and adding work without adding resources when updating a supported language.

[0006]

SUMMARY OF THE INVENTION According to the present invention, there is provided a telephone voice interpreting method for identifying a word spoken by a user inputted from a telephone network and outputting the same as recognition data; It is characterized by having means for translating into the language of the other party and outputting as language data, and means for outputting an audio signal synthesized corresponding to the language data.

Further, according to the telephone voice interpreting method of the present invention, a voice recognition device for identifying voice input from a telephone line and outputting character data determined according to the recognition data, and specifying the character data A translation device that converts the language into a language determined according to the selected communication partner and outputs the language data, and a voice synthesis device that outputs a voice signal synthesized according to the input language data; And performs translation between callers in different languages.

Further, the telephone voice interpreting method according to the present invention comprises a voice output device for outputting voice guidance to the telephone line for prompting the user of the incoming call from the telephone line to dial the telephone number of the communication partner. It is characterized by having.

Further, in the telephone voice interpreting method of the present invention, the dialing of the telephone number is transmitted by a PB signal, and a PB signal receiving apparatus determines the language of the country of the other party based on the telephone number of the received PB signal. It is characterized by doing.

[0010] In the telephone voice interpreting method according to the present invention, the guidance output by the voice output device is to store words spoken by a human in a memory in advance, and read out and output the words from the memory.

In the telephone speech interpreting method according to the present invention, the recognition information of the speech recognition device and the synthesis information of the speech synthesis device are collectively managed, and when the number of supported languages is increased, or when the existing recognition information and the existing synthesis information are combined. It is characterized by having control means for adding and changing only data at the time of information update or version upgrade.

[Operation] The telephone voice interpreting method of the present invention comprises:
This is a system that makes it possible to communicate even if the language of the other party cannot be spoken when making an international call, and has an interpreter that can translate the words spoken by the user into the language of the other party and listen. Further, since the interpreting device is incorporated in the telephone exchange network, the user does not need to purchase a new terminal. The present invention plays a role as an additional device of the telephone exchange network as shown in FIG. The present invention will be briefly described with reference to FIG.

(1). The user calls the interpreter 1. (2). Guidance for inquiring the telephone number of the other party is sent from the audio output device 6. (3).
The user dials the telephone number of the other party.
(4). From the communication device 1 (3). Set up a call for the telephone number entered in. (5) Notify the user that the call setting has been established with guidance. (6). A call is started between the user and the other party.

(7). The voice uttered by the user is passed through the voice recognition device 3 and converted into a character string. The character string is passed to the translation device 2. (8). The translation is performed by the translation device 2 into the language of the other party. (9). The translated character string is passed from the translation device 2 to the speech synthesis device 4. (10).
The speech is synthesized by the speech synthesizer 4 and transmitted to the other party.

[0015] Conversations from the other party to the user are also described in (7). ~ (10). The same operation as is performed.

Further, as the interpreting apparatus 1, voice recognition information and voice synthesis information are collectively managed by the control device 7. The voice recognition information and the voice synthesis information are managed by the control device 7 and the above procedure (3). By transmitting the speech recognition information and the speech synthesis information corresponding to the language of the country input in the above to the speech recognition device 3 and the speech synthesis device 4, the resources of the speech synthesis device 4 and the speech recognition device 3 are reduced, and the resources are reduced. Can be used effectively.

Further, even when the languages supported by the interpreter 1 increase, there is no need to extend the resources of the speech recognizer 3 and the speech synthesizer 4. Since the speech recognition information and the speech synthesis information are collectively managed by the control device 7, even when the corresponding language is increased, or when the existing speech recognition information and the speech synthesis information are updated or upgraded, the control device 7 can store the speech recognition information and the speech synthesis information. There is an advantage that it is only necessary to add and change only data.

Further, with respect to the audio output device 6, guidance to be transmitted is stored in advance as audio information.
In addition, a method of transmitting guidance using the voice synthesizing device 4 of the interpreting device 1 is also conceivable. However, in this method, there is a possibility that the guidance becomes a voice synthesized sound and sounds unnatural to the user.

[0019]

Embodiments of the present invention will now be described with reference to the drawings. FIG. 1 is a block diagram of an interpreter according to one embodiment of the present invention.

The interpreting apparatus 1 of this embodiment includes a translating apparatus 2, a speech recognizing apparatus 3, a speech synthesizing apparatus 4, a PB receiving apparatus 5, a speech output apparatus 6, a control apparatus 7, a switch (SW) 8, and
It includes a hub (HUB) 9.

As shown in the system block diagram of FIG. 2, the interpreter 1 is accommodated in a telephone network 20 and a telephone (Japan) 21
Alternatively, by accessing from the telephone (foreign) 22, different languages are automatically translated and transmitted from one to the other. 1, the speech recognition device 3 has a function of identifying a speech input from the telephone line 10 through the switch 8, converting the speech into character data, and transmitting the character data to the control device 7. The translation device 2 translates the character data from the speech recognition device 3 into a specified language and outputs it as language data.

The speech synthesizer 4 outputs a synthesized speech signal corresponding to the input language data. Audio output device 6
Means that voice guidance is transmitted through the switch 8 to the telephone line 10
It has the function of outputting to The voice guidance, which is uttered by a human, is stored in the memory of the apparatus 6 in advance. The PB receiving device 5 has a function of analyzing the received PB signal and transmitting it to the control device 7.

The control device 7 has a function of controlling the translation device 2, the voice recognition device 3, the voice synthesis device 4, the PB receiving device 5, the switch 8 and the hub 9, the function of setting a call, the voice recognition information and the voice synthesis information. And collectively manage the voice recognition device 3
And a function of reducing the resources of the speech synthesizer 4.

Further, even when the number of corresponding languages increases, there is no need to extend the resources of the speech recognition device 3 or the speech synthesis device 4. It has a function to add and change only necessary data in the memory of the control device when the number of supported languages increases or when existing speech recognition information and speech synthesis information are updated or upgraded.

The switch 8 is operated by an instruction from the control device 7.
It has a function of switching the connection with the telephone line 10 among the voice recognition device 3, voice output device 6, voice synthesis device 4, and PB receiving device 5. The hub 9 has a function of connecting the translation device 2, the speech recognition device 3, the speech output device 6, the speech synthesis device 4, the PB receiving device 5, and the control device 7 by Ethernet (registered trademark).

In this embodiment, the translation device 2, the speech recognition device 3, the speech output device 6, the speech synthesis device 4, and the PB receiving device 5 are defined as independent modules. An integrated module is also possible.

Next, the operation of this embodiment will be described. Step (abbreviated as S). In S1, when a call comes in from the telephone line 10, the control device 7 instructs the switch 8 to set a path to the audio output device 6 side. In S2, the switch 8 sets the path to the audio output device 6 according to the instruction of the control device 7. In S3, the switch 8 notifies the control device 7 that the setting of the path has been completed to the audio output device 6 side.

In S4, the control device 7 controls the sound output device 6
After receiving the notification that the path setting has been completed, the control unit instructs the voice output device 6 to output a guidance prompting the user to dial the telephone number (telephone number) of the other party. In S5, the audio output device 6 outputs guidance according to an instruction from the control device 7. Example: "Enter the telephone number of the other party." In S6, the audio output device 6 notifies the control device 7 that the output of the guidance has been completed. In S7,
After receiving the notification that the guidance output has ended, the control device 7 instructs the switch 8 to switch the path to the PB receiving device 5 side.

In S8, the switch 8 receives the instruction from the control device 7, and switches the path to the PB receiving device 5 side. S9
Then, the switch 8 notifies the control device 7 that the switching of the path is completed. In S10, the PB receiving device 5 recognizes the telephone number of the call partner input by the user and notifies the control device 7 and the translation device 2 of the telephone number. In S11, the control device 7
The data from the receiving device 5 is received, and a call is made to the input telephone number.

In S12, translation apparatus 2 determines the country code of the other party from the data of PB receiving apparatus 5, and determines the native language of that country. In S13, the speech recognition information and speech synthesis information of the native language determined by the translation device 2 are transmitted to the speech recognition device 3 and the speech synthesis device 4, respectively.
In S14, the speech recognition device 3 and the speech synthesis device 4
The translator 2 receives the speech recognition information and the speech synthesis information specified by the translator 2, and incorporates the speech recognition information and the speech synthesis information into each device.

In S15, the control device 7 instructs the switch 8 to switch the path to the audio output device 6 side. In S17, the switch 8 notifies the control device 7 that the switching of the path has been completed. In S18, the control device 7 receives the notification of the completion of the path switching, and instructs the audio output device 6 to output a guidance prompting to start a conversation. In S19, the voice output device 6 receives the instruction from the control device 7 and outputs guidance. Example: "Start a conversation. You can s
In "start talking" S20, the audio output device 6 notifies the control device 7 that the transmission of the guidance has been completed. In S21, after receiving the notification of the completion of the guidance transmission, the control device 7 instructs the switch 8 to switch the input path from the user to the voice recognition device 3 side. In S22, the switch 8 switches the path to the voice recognition device 3 in response to an instruction from the control device 7. In S23, the switch 8
Notifies the control device that the switching of the path to the voice recognition device 3 has been completed. In S24, the voice recognition device 3 recognizes the voice uttered by the user and notifies the translation device 2 through the control device 7 as character data. S2
In 5, the translation device 2 translates the character data received from the speech recognition device 3 into the language of the other party and notifies the control device 7.

In S26, the control device 7 instructs the switch 8 to switch the output path to the communication partner to the voice synthesizing device 4 side. In S27, the switch 8 switches the path to the voice synthesizing device 4 in response to an instruction from the control device 7. In S28, the switch 8 notifies the control device 7 that the path switching has been completed. In S29,
The control device 7 transmits the data translated by the translation device 2 to the speech synthesis device 4. In S30, the speech synthesizer 4
The speech data is synthesized from the translation data received from the translation device 2 and output to the other party.

After S31, S21 to S31 are repeated until the call ends. In addition, the same operation as S21 to S31 is performed for the voice from the other party.

[0034]

As described above, the first effect of the present invention is that communication between persons having different languages can be easily communicated by an international telephone.

The second effect is that the speech recognition information / speech synthesis information is collectively managed by the control device, the resources of the speech synthesis device / speech recognition are reduced, and the resources can be used effectively. Further, even when the number of supported languages increases, there is no need to extend the resources of the speech recognition device / speech synthesis device.

Further, since the voice recognition information and voice synthesis information are collectively managed by the control device, the data in the control device can be used even when the corresponding language is increased or when the existing voice recognition information / voice synthesis information is updated or upgraded. There is a merit that only addition / change is required.

[Brief description of the drawings]

FIG. 1 is a block diagram of an interpreter according to an embodiment of the present invention.

FIG. 2 is a block diagram illustrating a system according to the present exemplary embodiment.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 interpreter 2 translator 3 voice recognizer 4 voice synthesizer 5 PB receiver 6 voice output 7 controller 8 switch 9 hub

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI theme coat ゛ (reference) H04M 3/50 G10L 3/00 551C 551A F term (reference) 5B091 AA01 BA03 CB12 CB32 CD03 5D015 AA06 HH23 KK03 KK04 KK04 LL06 LL12 5D045 AB03 AB04 5K015 AA06 AA07 AA10 AF03 GA00 GA07 5K024 AA71 BB01 BB02 BB04 BB07 CC01 DD01 DD02 EE01 EE09 FF06 GG03

Claims (6)

[Claims] 1. A means for identifying a word spoken by a user input from a telephone network and outputting the same as recognition data, a means for translating the recognition data into a language of a designated communication partner and outputting the data as language data, Means for outputting a speech signal synthesized according to the language data. 2. A speech recognition device for recognizing speech input from a telephone line and outputting character data determined according to the recognition data, and a character recognition device which determines the character data according to a designated communication partner. And a speech synthesizer that outputs a speech signal synthesized corresponding to the input language data,
2. The method according to claim 1, further comprising the step of connecting to a telephone network to perform translation between callers of different languages. 3. The voice output device according to claim 2, further comprising a voice output device for outputting voice guidance to the telephone line to prompt the user receiving the call from the telephone line to dial the telephone number of the communication partner. Interpretation of telephone voice. 4. The method according to claim 3, wherein the dialing of the telephone number is transmitted by a PB signal, and a PB signal receiving apparatus determines the language of the country of the called party based on the telephone number of the received PB signal. Interpretation of telephone voice. 5. The telephone voice interpreting method according to claim 3, wherein the guidance output by the voice output device is to store in advance a word spoken by a human in a memory and read out and output from the memory. 6. The recognition information of the speech recognition device and the synthesis information of the speech synthesis device are collectively managed, and only data is used when the number of supported languages increases or when the existing recognition information and the synthesis information are updated or upgraded. 3. The telephone voice interpreting method according to claim 2, further comprising control means for performing additional changes.