Methods and apparatus for automatically extending the voice vocabulary of mobile communications devices

• US 20060173683A1
• Filed: 02/03/2006
• Published: 08/03/2006
• Est. Priority Date: 02/03/2005
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Voice Signal Technologies Incorporated (Microsoft Corporation)
• Inventors: Cohen, Jordan, Roth, Daniel L., Behrakis, Elias C.
• Granted Patent: US 8,160,884 B2
• Time in Patent Office: Days
• US Class Current: 704/251
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A method of improving the performance of a speech recognizer, the method involving: providing a lexicon for the speech recognizer; monitoring a user'"'"'s interaction with a network; accessing a plurality of words associated with the monitored interaction; and including the plurality of words in the lexicon.

Claims

1. A method of improving the performance of a speech recognizer, the method comprising:

providing a lexicon for the speech recognizer;

monitoring a user'"'"'s interaction with a network;

accessing a plurality of words associated with the monitored interaction; and

modifying the lexicon based on at least one of the plurality of words to improve the performance of the speech recognizer.

2. The method of claim 1, wherein the speech recognizer is on a mobile device.

3. The method of claim 2, wherein accessing comprises downloading the plurality of words from a server to the mobile device.

4. The method of claim 1, wherein the speech recognizer is located on a device connected to a network.

5. The method of claim 4, wherein the network is the Internet.

6. The method of claim 4, wherein the network is a telephone network.

7. The method of claim 1, wherein monitoring a user'"'"'s interaction with a network comprises monitoring the user'"'"'s interaction with a Web browser, and wherein the accessed plurality of words are associated with the user'"'"'s interaction with the Web browser.

8. The method of claim 7, wherein the plurality of words are obtained from at least one of queries, search results, and Web pages accessed as a result of the user'"'"'s interaction with the Web browser.

9. The method of claim 1, wherein monitoring a user'"'"'s interaction with a network comprises monitoring the user'"'"'s voice communication with one or more other parties via a telephone network, and wherein the accessed plurality of words are associated with the user'"'"'s voice communication with the one or more other parties via the telephone network.

10. The method of claim 1, wherein accessing comprises accessing the plurality of words dynamically.

11. The method of claim 1, wherein accessing comprises accessing the plurality of words at periodic intervals.

12. The method of claim 1, wherein accessing comprises accessing the plurality of words upon receipt of input from the user.

13. The method of claim 1, wherein accessing comprises accessing the plurality of words upon an occurrence of an event.

14. The method of claim 1, wherein the speech recognizer is biased in a manner that is based on a frequency of occurrence in the monitored interaction of words among the plurality of words.

15. The method of claim 1, wherein monitoring further comprises recording a classification of each of the plurality of words, and wherein the speech recognizer is biased in a manner that is based on the classification of each of the plurality of words.

16. The method of claim 1, wherein monitoring further comprises recording a usage context of each of the plurality of words, and wherein the speech recognizer is biased in a manner that is based on the usage context associated with each of the plurality of words.

17. A method implemented on a mobile device, the method comprising:

providing a lexicon for the speech recognizer;

monitoring a user'"'"'s interaction with a network;

capturing a plurality of words associated with the monitored interaction; and

adding the captured plurality of words to the lexicon.

18. The method of claim 17, wherein monitoring a user'"'"'s interaction with a network comprises monitoring the user'"'"'s interaction with a Web browser, and wherein the captured plurality of words are associated with the user'"'"'s interaction with the Web browser.

19. A method of extending a speech vocabulary on a mobile device having a speech recognizer, the method comprising:

storing on the mobile device a lexicon for the speech recognizer;

storing on the mobile device a plurality of words, the plurality of words being associated with a mobile user'"'"'s interaction with a web browser resident on the mobile device; and

importing the plurality of words into the lexicon.

20. The method of claim 19, wherein the plurality of words includes search terms entered into the browser by the mobile user and words contained within results returned by searches conducted using the search terms.

21. A method of extending a speech vocabulary on a mobile device having a speech recognizer, the method comprising:

storing on the mobile device a lexicon for the speech recognizer, the lexicon including a first plurality of words;

storing on the mobile device a second plurality of words and, for each word of the second plurality of words, storing a usage context associated with that word, the second plurality of words being associated with an application other than the speech recognizer; and

importing the second plurality of words and their associated usage contexts into the lexicon.

22. The method of claim 21, wherein at least one of the second plurality of words is associated with a usage context of a proper name.

23. The method of claim 21, wherein the speech recognizer is biased towards selecting candidate words from the lexicon that have a usage context that corresponds to the usage context of a word to be recognized by the speech recognizer.

24. A method of extending a speech vocabulary on a mobile device having a speech recognizer, the method comprising:

storing on the mobile device a lexicon for the speech recognizer, the lexicon including a plurality of words;

storing a second plurality of words on the mobile device, the second set of text words being associated with an application other than the speech recognizer;

for each word in the second plurality of words, recording a frequency of use of the word; and

importing the second plurality of words and their associated usage frequencies from the stored set of text words into the lexicon.

25. The method of claim 24, wherein the speech recognizer selects candidate words from the lexicon based at least in part on the candidates'"'"' usage frequency.