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

• US 8,160,884 B2
• Filed: 02/03/2006
• Issued: 04/17/2012
• 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: Roth, Daniel L., Cohen, Jordan, Behrakis, Elias P.
• Primary Examiner(s): Opsasnick, Michael N
• Application Number:
  US11/347,666
• Publication Number:
  US 20060173683A1
• Time in Patent Office: 2,265 Days
• Field of Search: 704/10, 704/251, 704/270.1
• US Class Current: 704/270.1
• 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

The invention is a method of improving the performance of a speech recognizer. The method generally involves: 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 on a mobile device which enables a user to interact with a communications network, the method comprising:

storing on the device a lexicon for the speech recognizer;

monitoring a user'"'"'s interaction over the network and capturing a plurality of words from the monitored interaction;

storing on the device the captured plurality of words; and

accessing the stored captured plurality of words and adding at least some of the captured plurality of words to the lexicon to improve the performance of the speech recognizer.

2. The method of claim 1, wherein monitoring the user'"'"'s interaction with a network comprises monitoring the user'"'"'s interaction with a Web browser, and wherein the captured plurality of words are taken from the user'"'"'s interaction with the Web browser.

3. The method of claim 2, wherein the captured 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.

4. The method of claim 1, wherein monitoring the 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 captured plurality of words are taken from the user'"'"'s voice communication with the one or more other parties via the telephone network.

5. The method of claim 1, wherein accessing the stored plurality of words is done dynamically.

6. The method of claim 1, wherein accessing the stored plurality of words is done at periodic intervals.

7. The method of claim 1, wherein accessing the stored plurality of words occurs in response to receiving an instruction from the user.

8. The method of claim 1, wherein accessing the stored captured plurality of words occurs in response to an occurrence of an event.

9. 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.

10. 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.

11. 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.

12. The method of claim 1, wherein adding at least some of the captured plurality of words to the lexicon comprises generating phonetic representations of the at least some of the captured plurality of words and adding the phonetic representations of the at least some of the captured plurality of words to the lexicon.