Granted Patent B2

US 8,401,853 · App. 12/888,012 · Granted Mar 19, 2013

Abstract

Claims (29)

1. A method of enhancing voice-enabled search, the method comprising:

recognizing, via a processor, received speech from a speaker to generate recognized speech;

identifying metadata about the speaker from the received speech; and

feeding the recognized speech and the metadata to a question-answering engine which selects and outputs at least one response to the speaker based on the recognized speech and metadata.

2. The method of claim 1 , wherein the metadata comprises demographic features.

3. The method of claim 2 , wherein the demographic features comprise at least one of age, gender, socio-economic group, nationality, and region.

4. The method of claim 1 , wherein identifying the metadata about the speaker is based on voice characteristics of the received speech.

5. The method of claim 1 , wherein recognizing the received speech and identifying the metadata occur in parallel.

6. The method of claim 1 , wherein the metadata identified about the speaker from the received speech supplements self-reported speaker demographic information.

7. A system comprising:

a processor;

a first module controlling the processor to receive a speech query from a user;

a second module controlling the processor to identify demographic information about the user from the speech query;

a question-answering engine controlling the processor to generate an answer to the speech query based on the demographic information.

8. The system of claim 7 , wherein the second module operates in parallel with a speech recognition engine.

9. The system of claim 7 , wherein the question-answering engine receives speech recognition output from the speech recognition engine.

10. The method of claim 7 , wherein the demographic information comprises at least one of age, gender, socio-economic group, nationality, and region.

11. The system of claim 7 , wherein identifying the demographic information about the user is based on voice characteristics of the received speech.

12. The system of claim 7 , wherein demographic information identified about the user from the received speech supplements self-reported user demographic information.

13. A non-transitory computer-readable storage medium storing instructions which, when executed by a computing device, cause the computing device to generate answers to a spoken query, the instructions comprising:

receiving a spoken query from a user;

identifying demographic metadata from the spoken query in parallel with speech recognition of the spoken query; and

passing the demographic metadata to a question-answering engine that also receives speech recognition output of the spoken query and generates at least one answer in response to the spoken query.

14. The non-transitory computer-readable storage medium of claim 13 , wherein the question-answering engine further outputs the at least one answer to the user.

15. The non-transitory computer-readable storage medium of claim 13 , wherein the demographic metadata comprises demographic features.

16. The non-transitory computer-readable storage medium of claim 15 , wherein the demographic features comprise at least one of age, gender, socio-economic group, nationality, and region.

17. The non-transitory computer-readable storage medium of claim 13 , wherein identifying the demographic metadata about the user is based on voice characteristics of the received speech.

18. The non-transitory computer-readable storage medium of claim 13 , wherein recognizing the received speech and identifying the metadata occur in parallel.

19. The non-transitory computer-readable storage medium of claim 13 , wherein the metadata identified about the user from the received speech supplements self-reported user demographic information.

Assignments (16)