JPS6128128A - Electronic translating device - Google Patents

Abstract

PURPOSE:To attain natural conversation between talkers each using a different language by identifying the distinction of sex from the input voice of the talkers and outputting the voice coincident with the distinction of sex. CONSTITUTION:When a voice of a female talker A is inputted to an electronic translator 1, the voice is led to voice recognition sections 1a, 1d, and after the recognition and translation of voice are executed respectively by the recognition section 1a and a translation section 1b, the result is fed to a voice synthesis/ switching section 1c. On the other hand, the recognition section 1d outputs an identification signal of the female voice to the switch section 1c. Thus, the switch section 1c synthesizes the result of translation in female voice and transmits the result to a talker B. When the talker B is a male, a male voice is inputted to the electronic translator 2 and the voice of the talker A is translated in a male voice similarly.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to an electronic interpreting device that enables conversations between speakers of different languages.

[Prior art]

Generally, this type of electronic interpreting device has the functions of speech recognition, speech translation, and speech synthesis.

However, the output of conventional electronic interpretation equipment for interpreting speech synthesis is fixed to either a male or female voice, regardless of the gender of the input speaker. There are cases where the gender does not match, which causes inconveniences in use, such as making the recipient feel uncomfortable.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide an electronic interpreting device in which the gender of the input voice and the gender of the output voice match, so that the receiver does not feel uncomfortable.

[Structure of the invention]

The electronic interpreting device according to the present invention includes means for identifying the gender of the input voice and means for switching the gender of the output voice to match the identified gender, thereby inputting the gender of the output voice. This is done to match the gender of the voice.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of an electronic interpreting system including an electronic interpreting device according to an embodiment of the present invention.

This system consists of two electronic interpreting devices 1 according to this embodiment.
and 2, and speaker A has different languages from each other.
, B enables conversation between them.

Here, the electronic interpreting device 1 includes a speech recognition section 1&, a speech translation section 1b, a speech synthesis/switching section 1C, and a speech identification section 1d. The speech recognition section 1a grammatically decomposes and recognizes the input speech from the speaker A, and converts this into an electrical signal to the speech translation section 1.
b (=guidance. The voice a station section 1b translates the above-mentioned recognized voice signal into a predetermined language and sends it to the voice synthesis/switching section 1C (=
Output.

The voice identification unit 1d receives the input voice from the speaker A, identifies whether it is a male voice or a female voice based on the voice characteristics, and sends a male voice/female voice identification signal to the voice synthesis/switching unit 1cl2. Send. The speech synthesis/switching section 1C synthesizes and outputs the translation result of the speech translation section 1b, but at this time, it switches between a male voice and a female voice in response to the above-mentioned identification signal and transmits it to the speaker B.

Note that the other electronic interpreting device 2 also has a voice recognition section 2 &
It is composed of a speech translation section 2b, a speech synthesis/switching section 2C, and a speech identification section 2d, and performs the same processing as described above.

Here, the specific operation of this electronic interpretation system will be explained using an example. For example, if the gender of the speaker is female, the input audio of a female voice will be
This input voice is guided to a voice recognition section 1a and a voice discrimination section 1d inside the device 1.

This input voice is processed by the voice recognition unit 1& and the voice translation unit 1bζ
Second, after the speech is recognized and translated as described above, it is guided to the speech synthesis/switching section 1c.

On the other hand, the voice identification section 1d identifies that the input voice is a female voice, and sends out a female voice identification signal to the voice synthesis/switching section 1cζ2. Thereby, the speech synthesis/switching unit 1c synthesizes a female voice based on the above translation result and transmits it to the speaker B. In other words, since the speaker is a woman, the voice interpreted as a female voice is output to speaker B.

Furthermore, if speaker B is a man, the second male voice is input to the other electronic interpreting device 2, and speaker A is similarly interpreted this time using a male voice.

〔Effect of the invention〕

As explained above, the present invention makes it possible to identify the gender of the speaker's input voice and produce an output voice that matches the gender, so that the listener can This has an extremely excellent effect of allowing natural dialogue to occur between speakers who speak in tongues.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an electronic interpreting system comprising two electronic interpreting devices according to an embodiment of the present invention. 1.2...Electronic interpreting devices 1a, 2a...Speech recognition unit 1b+2b ...Speech a station section 1a, 2c...Speech synthesis/switching section 1a, 2a...Speech identification section

Claims (1)

[Claims] In an electronic interpreting device that translates and outputs input voice into voice in a different language, means for identifying the gender of the input voice, and means for translating the gender of the input voice to match the identified gender. An electronic interpreting device characterized by comprising means for switching the gender of voice.