Communication System Using Mixed Translating While in Multilingual Communication

• US 20080071518A1
• Filed: 05/18/2007
• Published: 03/20/2008
• Est. Priority Date: 05/18/2006
• Status: Active Grant

• Current Assignee: University of Southern California
• Original Assignee: University of Southern California
• Inventors: Narayanan, Shrikanth, Georgiou, Panayiotis
• Granted Patent: US 8,706,471 B2
• Time in Patent Office: Days
• US Class Current: 704/2
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/55 Rule-based translation
• 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 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue

Abstract

A translation between a source language and a target language. The items are divided, with secondary source language items or named entities being identified. Those entities are translated in a different way. For example, they may be copied into the target language, or translated in a special way that is based on their meaning, e.g., into a term that has a more descriptive meaning in the target language.

Claims

1. A method, comprising:

receiving information to be translated;

identifying at least one part of the information which needs to be translated differently within the information to be translated; and

translating the at least one part in a different way.

2. A method as in claim 1, wherein the at least one part is a word representing a named entity.

3. A method as in claim 2, wherein the information includes a primary source language, and said at least one part is a word in a secondary source language.

4. A method as in claim 1, wherein the information includes a primary source language, and said at least one part is a word in a secondary source language.

5. A method as in claim 1, wherein said information to be translated is primarily in a primary source language, but also includes words in a secondary source language, and wherein said identifying comprises identifying words in the secondary source language, and wherein said translating comprises reproducing the words in the secondary language into a target language, without translating.

6. A method as in claim 1, wherein said at least one part includes proper names.

7. A method, comprising:

receiving information to be translated in a source language, to be translated to a target language, wherein said information includes a first part that is in a primary source language and a second part that is in a secondary source language;

segmenting said information to include a distinction between said primary source language parts and said secondary source language parts;

translating said primary source language parts into the target language; and

carrying out an operation on the secondary source language that is different than said translating.

8. A method as in claim 7, wherein said operation comprises copying the secondary source language parts into the target language without modification.

9. A method as in claim 7, wherein said operation comprises translating the secondary source language parts using a different translation that is specific for second source language parts.

10. A method as in claim 7, further comprising detecting named entities as parts in said secondary source language.

11. A method as in claim 10, wherein said named entities include proper names.

12. A method as in claim 8, wherein said copying comprises using an automated speech reading part to read the speech into the target language.

13. A method as in claim 8, wherein said copying comprises recording the speech in the source language, and replaying the speech in the target language.

14. A method as in claim 11, wherein said detecting named entities further comprises determining if the named entity has a translation in the target language, and if not then copying the information into the target language.

15. A method as in claim 14, wherein said named entity is a word that has multiple translations into the target language, and has a different translation when used as a named entity than it has when used as a non-named entity.

16. A method, comprising:

obtaining information in a source language to be translated;

segmenting said information into a first part that represents information in the source language, and a second part that represents named entities in the source language;

determining if the named entities in the source language have a different translation into the target language than they would have if non-named entities; and

translating said named entities using said different translation.

17. A method as in claim 16, wherein said named entities are in a primary source language.

18. A method as in claim 16, wherein the named entities are in a secondary source language.

19. A computer system, comprising:

a first part, that receives information in a source language to be translated into a target language;

a processor, programmed to divide the information to be translated into a first part which is transmitted according to a first translation between said source language and said target language, and a second part which is translated in a special way that is different than the first translation, wherein the special way translates the second part in a first case by copying the second part into the target language, and in a second case by translating the second part into the target language, and wherein the processor converts the information from the source language into the target language.

20. A system as in claim 19, wherein said special way translates into a generic term in the target language.