SYSTEM AND METHOD FOR LANGUAGE TRANSLATION IN A HYBRID PEER-TO-PEER ENVIRONMENT

• US 20120078609A1
• Filed: 09/24/2010
• Published: 03/29/2012
• Est. Priority Date: 09/24/2010
• Status: Active Grant

• Current Assignee: Damaka, Inc.
• Original Assignee: Damaka, Inc.
• Inventors: CHATURVEDI, SIVAKUMAR, GUNDABATHULA, SATISH
• Granted Patent: US 8,468,010 B2
• Time in Patent Office: Days
• US Class Current: 704/3
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• 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 15/005 Language recognition

Abstract

An improved system and method are disclosed for peer-to-peer communications. In one example, the method enables an endpoint to send and/or receive audio speech translations to facilitate communications between users who speak different languages.

Claims

1. A method for communicating, by a first endpoint, in an ongoing peer-to-peer communication session between the first endpoint and a second endpoint in a hybrid peer-to-peer network comprising:

establishing, by the first endpoint, a communications route directly between the first endpoint and the second endpoint, wherein signaling communications are sent directly from the first endpoint to the second endpoint via a signaling path provided by the communications route;

receiving, by the first endpoint, audio speech input in a first spoken language from a user of the first endpoint;

determining, by the first endpoint, whether the audio speech input is to be translated from the first spoken language to a second spoken language;

sending, by the first endpoint, the audio speech input to a language translation component accessible to the first endpoint via the hybrid peer-to-peer network if the audio speech input is to be translated from the first spoken language to the second spoken language, wherein the first endpoint does not send the audio speech input directly to the second endpoint if the audio speech input is to be translated from the first spoken language to the second spoken language; and

sending, by the first endpoint, the audio speech input directly to the second endpoint via the communications route if the audio speech input is not to be translated from the first spoken language to the second spoken language.

2. The method of claim 1 further comprising sending, by the first endpoint, non-audio data directly to the second endpoint via the communications route regardless of whether the audio speech input is to be translated from the first spoken language to the second spoken language.

3. The method of claim 1 further comprising:

receiving, by the first endpoint, a translation of the audio speech input from the language translation component; and

providing the translation of the audio speech input to the user of the first endpoint.

4. The method of claim 1 wherein the sending the audio speech input to the language translation component includes:

establishing, by the first endpoint, a signaling path with a speech-to-text module of the language translation component; and

sending the audio speech input to the speech-to-text module.

5. The method of claim 1 further comprising:

performing, by the first endpoint, an authentication process with an access server in the hybrid peer-to-peer network; and

receiving, by the first endpoint, a profile from the access server following the authentication process, wherein the profile identifies the second endpoint as an endpoint with which the first endpoint has permission to communicate and identifies that the second endpoint is associated with the second spoken language.

6. The method of claim 5 further comprising receiving, by the first endpoint, a list of languages from the access server, wherein the list of languages includes the first and second spoken languages and identifies spoken languages that can be translated by the language translation component.

7. The method of claim 1 further comprising receiving, by the first endpoint, a notification directly from the second endpoint that the second endpoint is associated with the second spoken language.

8. The method of claim 1 further comprising:

receiving, by the first endpoint from the language translation component, audio speech originating from the second endpoint, wherein the audio speech originating from the second endpoint is translated from the second spoken language to the first spoken language by the language translation component before being received by the first endpoint; and

producing, by the first endpoint, the audio speech received from the language translation component as an audible sound.

9. The method of claim 1 further comprising:

establishing, by the first endpoint, a second communications route directly between the first endpoint and a third endpoint, wherein signaling communications are sent directly from the first endpoint to the third endpoint via a signaling path provided by the second communications route;

identifying, by the first endpoint, that the third endpoint is associated with the first spoken language; and

sending, by the first endpoint, the audio speech input directly to the third endpoint via the second communications route.

10. A method for communicating, by a first endpoint in a hybrid peer-to-peer network, in an ongoing communication session with second and third endpoints via a bridge comprising:

identifying, by the first endpoint, that the first endpoint is associated with a first spoken language, the second endpoint is associate with a second spoken language, and the third endpoint is associated with a third spoken language;

sending, by the first endpoint, a request to the bridge for an inbound port and an outbound port to be provisioned on the bridge for the first endpoint;

notifying, by the first endpoint, a language translation component in the hybrid peer-to-peer network of the inbound port, wherein the language translation component is accessible to the first endpoint via the hybrid peer-to-peer network, and wherein the notifying instructs the language translation component to send audio received from the first endpoint to the inbound port;

sending to the language translation component, by the first endpoint, audio speech input received by the first endpoint from a user of the first endpoint, wherein the audio speech input sent by the first endpoint is in the first spoken language; and

receiving, by the first endpoint, audio speech from the second and third endpoints directly from the outbound port on the bridge, wherein the audio speech received by the first endpoint via the outbound port was sent by the second and third endpoints in the second and third spoken languages, respectively, and wherein the audio speech received by the first endpoint directly from the outbound port is received in the first spoken language.

11. The method of claim 10 further comprising:

identifying, by the first endpoint, that a fourth endpoint is associated with the first spoken language; and

sending, by the first endpoint, the audio speech received from the user of the first endpoint directly to the fourth endpoint without using the bridge.

12. The method of claim 10 further comprising:

identifying, by the first endpoint, that a fourth endpoint is associated with the first spoken language; and

sending, by the first endpoint, the audio speech received from the user of the first endpoint to the inbound port on the bridge.

13. The method of claim 10 further comprising sending, by the first endpoint, non-audio data directly to at least one of the second and third endpoints, wherein the non-audio data does not pass through the bridge.

14. The method of claim 10 further comprising receiving, by the first endpoint, non-audio data directly from at least one of the second and third endpoints, wherein the non-audio data does not pass through the bridge.

15. The method of claim 10 further comprising directly notifying the second and third endpoints, by the first endpoint, of the inbound and outbound ports, wherein the notifying does not use the bridge.

16. The method of claim 10 further comprising receiving, by the first endpoint, notifications directly from each of the second and third endpoints of inbound and outbound ports corresponding to the second and third endpoints.

17. The method of claim 10 further comprising sending an indication, by the first endpoint, to the language translation component that the audio speech input is to be translated into the second and third languages.

18. A method for translating audio speech in a hybrid peer-to-peer network comprising:

receiving, by a speech-to-text module, first audio speech media from a first endpoint via a hybrid peer-to-peer network, wherein the first audio speech media is in a first human language;

converting, by the speech-to-text module, the first audio speech media into original text;

sending, by the speech-to-text module, the original text to a language translation module;

translating, by the language translation module, the original text into translated text in a second human language;

sending, by the language translation module, the translated text to a text-to-speech module;

converting, by the text-to-speech module, the translated text into second audio speech media, wherein the second audio speech media is in the second human language; and

sending the second audio speech media to a second endpoint in the hybrid peer-to-peer network.

19. The method of claim 18 wherein sending the second audio speech media is performed by the text-to-speech module.

20. The method of claim 18 further comprising sending, by the speech-to-text module, the translated text to the first endpoint.

21. The method of claim 20 further comprising waiting, by the speech-to-text module, for an approval of the translated text from the first endpoint before sending the original text to the language translation module.

22. The method of claim 18 wherein sending the second audio speech media to the second endpoint includes sending the second audio speech media to a port on a bridge identified by the first endpoint.

23. The method of claim 22 further comprising:

translating, by the language translation module, the original text into translated text in a third human language;

sending, by the language translation module, the translated text to the text-to-speech module;

converting, by the text-to-speech module, the translated text into third audio speech media, wherein the third audio speech media is in the third human language; and

sending the third audio speech media to a third endpoint by sending the third audio speech media to a port on the bridge identified by the first endpoint.

24. The method of claim 23 wherein the ports for the second and third audio speech media are the same port.

25. The method of claim 18 further comprising receiving, by the speech-to-text module, instructions from the first endpoint identifying the first and second human languages.

26. An endpoint device comprising:

a network interface;

a processor coupled to the network interface; and

a memory coupled to the processor and containing a plurality of instructions for execution by the processor, the instructions including instructions for;

performing an authentication process with an access server in a hybrid peer-to-peer network, wherein the authentication process authorizes the first endpoint to access the hybrid peer-to-peer network;

receiving a profile from the access server identifying a second endpoint as an endpoint within the hybrid peer-to-peer network with which the first endpoint has permission to communicate;

determining that a user of the first endpoint has designated a first human language to be used by the first endpoint;

establishing a communications route directly between the first endpoint and the second endpoint, wherein signaling communications are sent directly from the first endpoint to the second endpoint via the communications route;

determining that the second endpoint has designated a second human language to be used by the second endpoint;

receiving audio speech input in the first human language from the user of the first endpoint; and

sending the audio speech input to a language translation module for translation to the second human language.

27. The endpoint device of claim 26 further comprising instructions for sending non-audio data directly to the second endpoint via the communications route.

28. The endpoint device of claim 26 wherein the instructions for determining that the second endpoint has designated that the second human language be used by the second endpoint include instructions for obtaining an identifier representing the second human language from the profile.

29. The endpoint device of claim 26 wherein the instructions for determining that the second endpoint has designated that the second human language be used by the second endpoint include instructions for obtaining an identifier representing the second human language directly from the second endpoint.