Method for speech interpretation service and speech interpretation server

• US 20020046035A1
• Filed: 03/20/2001
• Published: 04/18/2002
• Est. Priority Date: 10/17/2000
• Status: Active Grant

• Current Assignee: Rakuten Group, Inc.
• Original Assignee: Hitachi, Ltd.
• Inventors: Obuchi, Yasunari, Koizumi, Atsuko, Kitahara, Yoshinori, Mizutani, Seiki
• Granted Patent: US 7,130,801 B2
• Time in Patent Office: Days
• US Class Current: 704/277
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding

Abstract

A speech interpretation server, and a method for providing a speech interpretation service, are disclosed. The server includes a speech input for receiving an inputted speech in a first language from a mobile terminal, a speech recognizer that receives the inputted speech and converts the inputted speech into a prescribed symbol string, a language converter that converts the inputted speech converted into the prescribed symbol string into a second language, wherein the second language is different from the first language, and a speech output that outputs the second language to the mobile terminal. The method includes the steps of providing an interpretation server having resident thereon a plurality of registered sentences to be interpreted, activating a translation connection between the mobile terminal and the interpretation server, receiving speech, in a first language, inputted to the mobile terminal via the translation connection, at the interpretation server, recognizing and interpreting the speech inputted based on a comparison of the inputted speech to the plurality of registered sentences to be interpreted, and outputting a translation correspondent to the second language.

Claims

1. A method for providing a speech interpretation service, comprising:

providing an interpretation server having resident thereon a plurality of registered sentences to be interpreted;

communicating to a terminal communicatively connected to the interpretation server, at least one of the plurality of registered sentences to be interpreted;

receiving speech, in a first language, inputted to the mobile terminal, at the interpretation server;

recognizing the speech inputted based on a comparison of the inputted speech to the plurality of registered sentences to be interpreted;

interpreting, by the interpretation server, the recognized speech into a second language, according to said recognizing; and

outputting a translation signal correspondent to the second language to the terminal from the interpretation server.

2. The method of claim 1, wherein the communicative connection comprises a mobile internet network, further comprising:

displaying an interpretable language classification menu on the mobile terminal prior to said receiving speech in a first language;

receiving, from the mobile terminal, a selection input of the first language and the second language from the language classification menu.

3. The method of claim 1, wherein the registered sentences comprise a plurality of scenes, each scne including therein a plurality of model sentences, further comprising:

displaying on the mobile terminal a model sentence field;

receiving a model sentence field selection of a model sentence from the model sentence field; and

wherein said interpreting comprises interpreting the inputted speech according to the model sentence field selection.

4. The method of claim 1, wherein the communicative connection is a telephonic audio network connection.

5. The method of claim 4, wherein the translation signal comprises an audio signal, wherein said outputting a translation signal correspondent to the second language to the mobile terminal comprises outputting the audio signal on the mobile terminal via the telephone network.

6. The method of claim 1, further comprising:

receiving an approval instruction from the mobile terminal of the second language before said outputting a translation signal correspondent to the second language.

7. The method of claim 6, wherein:

the approval instruction is an audio approval instruction given at the mobile terminal, and wherein the approval instruction is selected from a speaking set, and wherein the audio approval instruction is at least one selected from the group consisting of a specific spoken word, a specific spoken phrase, and a specific spoken sentence, from the speaking set and wherein said outputting a translation signal correspondent to the second language is in accordance with the approval instruction.

8. The method of claim 6, wherein:

the approval instruction is a press button approval instruction given at the mobile terminal, and wherein said outputting a translation signal correspondent to the second language is in accordance with the approval instruction.

9. The method of claim 7, further comprising repeating said outputting a translation signal correspondent to the second language in accordance with the approval instruction upon each receipt of the approval instruction.

10. The method of claim 1, further comprising:

narrowing, based on a dictionary database, the ones of the plurality of registered sentences for said recognizing the speech inputted based on a comparison of the inputted speech to the plurality of registered sentences to be interpreted.

11. The method of any one of claims 1, further comprising:

identifying the mobile terminal based on at least one identifying characteristic; and

charging a predetermined fee to the identified mobile terminal for said interpreting.

13. The speech interpretation server of claim 12, wherein said memory comprises:

a command sentence table, including a plurality of command sentences, wherein said speech recognizer differentiates the plurality of model sentences from the plurality of command sentences, and wherein each of the plurality of command functions instructs an action by the speech interpretation server.

14. The speech interpretation server of claim 12, further comprising:

a comparator, wherein said comparator compares the inputted speech to the plurality of model sentences in order to generate the prescribed symbol string.

15. The speech interpretation server of claim 14, wherein each of the plurality of model sentences is classified according to a scene of use.

16. The speech interpretation server of claim 12, wherein said speech output comprises a speech synthesizer output that outputs the second language to the mobile terminal in audio.

17. The speech interpretation server of claim 12, further comprising:

an authorizer, wherein said authorizer identifies the mobile terminal based on at least one identifying characteristic; and

a billing database, wherein the mobile terminal correspondent to the at least one identifying characteristic is billed by said billing database for use of the speech interpretation server for a predetermined time.

19. The speech interpretation service of claim 18, wherein said at least one connection is at least one selected from the group consisting of a mobile internet connection and a telephone network connection.

12. A speech interpretation server, comprising:

a speech input for receiving an inputted speech in a first language from a mobile terminal;

a speech recognizer that receives the inputted speech and converts the inputted speech into a prescribed symbol string;

a memory having stored thereon a plurality of model sentences, wherein the prescribed symbol string is present among the plurality of model sentences;

a language converter that converts the inputted speech converted into the prescribed symbol string into a second language, wherein the second language is different from the first language; and

a speech output that outputs the second language in audio to the mobile terminal.

18. A speech interpretation service, comprising:

a mobile terminal connected to a communication server, wherein the communication server comprises;

a speech input for receiving an inputted speech in a first language from said mobile terminal;

a speech recognizer that receives the inputted speech and converts the inputted speech into a prescribed symbol string;

a model sentence table for storing a plurality of model sentences, wherein the prescribed symbol string is present among the plurality of model sentences;

a language converter that converts the inputted speech converted into the prescribed symbol string into a second language, wherein the second language is different from the first language; and

a speech output that outputs the second language to said mobile terminal; and

at least one connection between said mobile terminal and the communication server.