SPEECH TRANSLATION SYSTEM, DICTIONARY SERVER, AND PROGRAM

• US 20120166176A1
• Filed: 03/03/2010
• Published: 06/28/2012
• Est. Priority Date: 07/16/2009
• Status: Active Grant

• Current Assignee: National Institute of Information and Communications Technology
• Original Assignee: National Institute of Information and Communications Technology
• Inventors: Nakamura, Satoshi, Sumita, Eiichiro, Ashikari, Yutaka, Kimura, Noriyuki, Hori, Chiori
• Granted Patent: US 9,442,920 B2
• Time in Patent Office: Days
• US Class Current: 704/7
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A conventional speech recognition dictionary, translation dictionary and speech synthesis dictionary used in speech translation have inconsistencies. Inconsistency among dictionaries can be solved with a dictionary server including an all-language group dictionary storage unit capable of having stored therein, for two or more languages, two or more all-language term information pieces for speech recognition, translation and speech synthesis, an information-for-speech-recognition sending unit that acquires information for speech recognition including speech recognition information of a term for all languages, and sends the information to one or more speech recognition servers, an information-for-translation sending unit that acquires information for translation including the written form of the term for all languages, and sends the information to one or more translation servers, and an information-for-speech-synthesis sending unit that acquires information for speech synthesis including speech synthesis information of a term for all languages, and sends the information to one or more speech synthesis servers.

Claims

1. A speech translation system including a dictionary server, one or more speech recognition servers, one or more translation servers, and one or more speech synthesis servers,

wherein the dictionary server comprises:

an all-language group dictionary storage unit capable of having stored therein two or more pieces of all-language term information, each piece of all-language term information including term information with respect to each of two or more languages, each term information including a written form of a term that has one or more words, speech recognition information for performing speech recognition on the term, and speech synthesis information for performing speech synthesis on the term, the term information being associated with each other with respect to all of the two or more languages;

an information-for-speech-recognition sending unit that acquires, from the all-language group dictionary storage unit, information for speech recognition that includes speech recognition information of the term with respect to all of the two or more languages, or at least two of the two or more languages, and sends the information for speech recognition to the one or more speech recognition servers;

an information-for-translation sending unit that acquires, from the all-language group dictionary storage unit, information for translation that includes the written form of the term with respect to all of the two or more languages, or at least two of the two or more languages, and sends the information for translation to the one or more translation servers; and

an information-for-speech-synthesis sending unit that acquires, from the all-language group dictionary storage unit, information for speech synthesis that includes speech synthesis information of the term with respect to all of the two or more languages, or at least two of the two or more languages, and sends the information for speech synthesis to the one or more speech synthesis servers,

each speech recognition server comprises;

an information-for-speech-recognition storage unit capable of having stored therein information for speech recognition with respect to all of the two or more languages, or at least two of the two or more languages;

an information-for-speech-recognition receiving unit that receives, from the dictionary server, information for speech recognition with respect to all of the two or more languages, or at least two of the two or more languages;

an information-for-speech-recognition accumulation unit that accumulates the information for speech recognition received by the information-for-speech-recognition receiving unit in the information-for-speech-recognition storage unit;

a speech information receiving unit that receives speech information that is information of speech input to a first terminal apparatus;

a speech recognition unit that performs speech recognition on the speech information received by the speech information receiving unit by using information for speech recognition in the information-for-speech-recognition storage unit, and acquires a speech recognition result; and

a speech recognition result sending unit that sends the speech recognition result,

each translation server comprises;

an information-for-translation storage unit capable of having stored therein information for translation with respect to all of the two or more languages, or at least two of the two or more languages;

an information-for-translation receiving unit that receives, from the dictionary server, information for sound translation with respect to all of the two or more languages, or at least two of the two or more languages;

an information-for-translation accumulation unit that accumulates the information for translation received by the information-for-translation receiving unit in the information-for-translation storage unit;

a speech recognition result receiving unit that receives the speech recognition result;

a translation unit that translates the speech recognition result received by the speech recognition result receiving unit into a target language by using information for translation in the information-for-translation storage unit, and acquires a translation result; and

a translation result sending unit that sends the translation result,

each speech synthesis server comprises;

an information-for-speech-synthesis storage unit capable of having stored therein information for speech synthesis with respect to all of the two or more languages, or at least two of the two or more languages;

an information-for-speech-synthesis receiving unit that receives, from the dictionary server, information for speech synthesis with respect to all of the two or more languages, or at least two of the two or more languages;

an information-for-speech-synthesis accumulation unit that accumulates the information for speech synthesis received by the information-for-speech-synthesis receiving unit in the information-for-speech-synthesis storage unit;

a translation result receiving unit that receives the translation result;

a speech synthesis unit that performs speech synthesis on the translation result received by the translation result receiving unit by using information for speech synthesis in the information-for-speech-synthesis storage unit, and acquires a speech synthesis result; and

a speech synthesis result sending unit that sends the speech synthesis result to a second terminal apparatus.

2. The speech translation system according to claim 1,

wherein the speech recognition unit of the speech recognition server comprises:

a speech recognition determination section that determines whether speech recognition processing performed on the speech information received by the speech information receiving unit has succeeded or failed;

a speech recognition information sending instruction section that issues, to the dictionary server, an instruction to send speech recognition information in a case where the speech recognition determination section has determined that the speech recognition processing has failed; and

a speech recognition section that performs speech recognition on the speech information received by the speech information receiving unit by using information for speech recognition in the information-for-speech-recognition storage unit, and acquires a speech recognition result, and also performs speech recognition by using speech recognition information that the information-for-speech-recognition receiving unit has received from the dictionary server in response to sending of the instruction, and acquires a speech recognition result, and

the information-for-speech-recognition receiving unit receives the speech recognition information from the dictionary server in response to sending of the instruction.

3. The speech translation system according to claim 1,

wherein the translation unit of the translation server comprises:

a translation determination section that determines whether translation processing performed on the speech recognition result received by the speech recognition result receiving unit has succeeded or failed;

a translation information sending instruction section that issues, to the dictionary server, an instruction to send the written form of the term in the target language in a case where the translation determination section has determined that the translation processing has failed; and

a translation section that translates the speech recognition result received by the speech recognition result receiving unit into the target language by using information for translation in the information-for-translation storage unit, and acquires a translation result, and also translates the speech recognition result into the target language by using the written form of the term in the target language that the information-for-translation receiving unit has received from the dictionary server in response to sending of the instruction, and acquires a translation result,

wherein the information-for-translation receiving unit receives the written form of the term in the target language from the dictionary server in response to sending of the instruction.

4. The speech translation system according to claim 1,

wherein the speech synthesis unit of the speech synthesis server comprises:

a speech synthesis determination section that determines whether speech synthesis processing performed on the translation result received by the translation result receiving unit has succeeded or failed;

a speech synthesis information sending instruction section that issues, to the dictionary server, an instruction to send speech synthesis information in a case where the speech synthesis determination section has determined that the speech synthesis processing has failed; and

a speech synthesis section that performs speech synthesis on the translation result received by the translation result receiving unit by using information for speech synthesis in the information-for-speech-synthesis storage unit, and acquires a speech synthesis result, and also performs speech synthesis on the translation result by using the speech synthesis information that the information-for-speech-synthesis receiving unit has received from the dictionary server in response to sending of the instruction, and acquires a speech synthesis result, and

the information-for-speech-synthesis receiving unit receives the speech synthesis information from the dictionary server in response to sending of the instruction.

5. The speech translation system according to claim 1,

wherein the dictionary server further comprises:

a written form acquiring unit that acquires a written form of a term that does not exist in the all-language group dictionary storage unit from web pages on one or more web servers on the Internet; and

a written form accumulation unit that accumulates the written form of the term acquired by the written form acquiring unit in the all-language group dictionary storage unit.

6. The speech translation system according to claim 5,

wherein the dictionary server further comprises:

an information accepting unit that accepts any information of the term information from one or more third terminal apparatuses; and

an information accumulation unit that accumulates the information accepted by the information accepting unit in the all-language group dictionary storage unit in association with the written form of the term corresponding to the information accepted by the information accepting unit in a language corresponding to the information accepted by the information accepting unit.

7. The speech translation system according to claim 5,

wherein the dictionary server further comprises an output unit that outputs the all-language term information or part of the all-language term information, and

in a case where the output unit outputs the all-language term information or part of the all-language term information, the output unit outputs the all-language term information or part of the all-language term information in visually different manners between a case where all predetermined information exists with respect to all of the two or more languages, and a case where part of the predetermined information is missing.

8. A dictionary server that constitutes the speech translation system of claim 1.

9. A speech recognition server that constitutes the speech translation system of claim 1.

10. A translation server that constitutes the speech translation system of claim 1.

11. A speech synthesis server that constitutes the speech translation system of claim 1.

12. A program for causing, while allowing storage in a storage medium of two or more pieces of all-language term information, each piece of all-language term information including term information with respect to each of two or more languages, each term information including a written form of a term that has one or more words, speech recognition information for performing speech recognition on the term, and speech synthesis information for performing speech synthesis on the term, and the term information being associated with each other with respect to all of the two or more languages, a computer to function as:

an information-for-speech-recognition sending unit that acquires, from the storage medium, information for speech recognition that includes speech recognition information of the term with respect to all of the two or more languages, or at least two of the two or more languages, and sends the information for speech recognition to the one or more speech recognition servers;

an information-for-translation sending unit that acquires, from the storage medium, information for translation that includes the written form of the term with respect to all of the two or more languages, or at least two of the two or more languages, and sends the information for translation to the one or more translation servers; and

an information-for-speech-synthesis sending unit that acquires, from the storage medium, information for speech synthesis that includes speech synthesis information of the term with respect to all of the two or more languages, or at least two of the two or more languages, and sends the information for speech synthesis to one or more speech synthesis servers.