Method and system for searching recorded speech and retrieving relevant segments

• US 7,039,585 B2
• Filed: 09/24/2001
• Issued: 05/02/2006
• Est. Priority Date: 04/10/2001
• Status: Expired due to Term

• Current Assignee: Uniloc 2017 LLC (FIG LLC (d/b/a Fortress Investment Group LLC))
• Original Assignee: International Business Machines Corporation
• Inventors: Kern, Robert, Wilmot, Gerald Johann
• Primary Examiner(s): ABEBE, DANIEL DEMELASH
• Application Number:
  US09/962,659
• Publication Number:
  US 20020147592A1
• Time in Patent Office: 1,681 Days
• Field of Search: 704/235, 379/88.14
• US Class Current: 704/235
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06F 16/685 using automatically derived transcript of audio data, e.g. lyrics speech recognition G10L15/00
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

A system and method for searching recorded speech is disclosed. The system and method comprises converting the recorded speech into text using a voice recognition system. As the speech is being converted, naturally occurring breaks in the languages will be used to take time indexes from the recording. The system and method includes creating a full text index of the recorded speech utilizing an information extender. The full text index contains a plurality of time stamps that point to the occurrence of words in the recorded speech. Finally, the text is searched by a full text search server that has linguistic search capabilities using the full text index. Finally, the searched text, the text index and the recorded speech are stored in the database. The recorded speech is searched by locating relevant phrases or words, and then mapping the time stamps associated with the relevant phrases words back to the recorded speech in the database.

Claims

1. A method for searching recorded speech in a database, the method comprising:

(a) converting the recorded speech to text utilizing a voice recognition system, wherein breaks in the recorded speech are used to provide a plurality of time stamps;

(b) creating a full text index of the recorded speech utilizing an information extender, the full text index containing the plurality of time stamps which point to the occurrence of words in the recorded speech;

(c) searching the text by a full text server utilizing the full text index; and

(d) storing the searched text, the full text index and the recorded speech in the database, wherein specific contents of the recorded speech can be played back utilizing the full text index without listening to the complete recording.

2. The method of claim 1 wherein the converting step (a) includes the step of (a1), storing the recorded speech in an index tool utilizing an audio extender.

3. The method of claim 1 wherein the plurality of time stamps are produced during step (a).

4. The method of claim 1 wherein the recorded speech is voicemail messages provided by an Internet-based voicemail system.

5. The method of claim 4 wherein the voicemail messages are played back by accessing the voicemails, saying the keywords and listening to the phrases that contain the key words.

6. The method of claim 1 wherein the recorded speech is voicemail messages provided by an Internet-based voicemail system.

7. The method of claim 6 wherein a voicemail server is accessed via the Internet, search words and/or phrases of the voicemail message, observing a summary of the search results and listening to the relevant portions of the messages.

8. The method of claim 1 wherein the recorded speech comprises a recorded meeting.

9. The method of claim 8 wherein a recorded meeting is played back by listening to relevant portions of the recorded meeting, typing words and/or phrases that are being looked for, observing a textual summary of what was found and listening to the relevant contents based upon the textual summary.

10. The method of claim 8 wherein the recorded meeting is played back by listening to relevant portions of the speech, saying the words and/or phrases that are being looked for and listening to the relevant contents.

11. The method of claim 1 wherein the recorded speech includes video.

12. The method of claim 11 wherein the video is searched based upon the full text index.

13. The method of claim 1 wherein the recorded speech comprises a news program.

14. A system for searching recorded speech in a database, the system comprising:

means for converting the recorded speech to text utilizing a voice recognition system, wherein breaks in the recorded speech are used to provide a plurality of time stamps;

means for creating a full text index of the recorded speech utilizing an information extender, the full text index containing the plurality of time stamps which point to the occurrence of words in the recorded speech;

means for searching the text by a full text server utilizing the full text index; and

means for storing the searched text, the full text index and the recorded speech in the database, wherein specific contents of the recorded speech can be played back utilizing the full text index without listening to the complete recording.

15. The system of claim 14 wherein the converting means includes the means for storing the recorded speech in an index tool utilizing an audio extender.

16. The system of claim 14 wherein the plurality of time stamps are produced by the converting means within.

17. The system of claim 14 wherein the recorded speech is voicemail messages provided by an Internet-based voicemail system.

18. The system of claim 17 wherein the voicemail messages are played back by accessing the voicemails, saying the keywords and listening to the phrases that contain the key words.

19. The system of claim 14 wherein the recorded speech is voicemail messages provided by an Internet-based voicemail system.

20. The system of claim 19 wherein a voicemail server is accessed via the Internet, search words and/or phrases of the voicemail message, observing a summary of the search results and listening to the relevant portions of the messages.

21. The system of claim 14 wherein the recorded speech comprises a recorded meeting.

22. The system of claim 21 wherein a recorded meeting is played back by listening to relevant portions of the recorded meeting, typing words and/or phrases that are being looked for, observing a textual summary of what was found and listening to the relevant contents based upon the textual summary.

23. The system of claim 21 wherein the recorded meeting is played back by listening to relevant portions of the speech, saying the words and/or phrases that are being looked for and listening to the relevant contents.

24. The computer readable medium of claim 23 wherein the recorded speech comprises a news program.

25. The system of claim 14 wherein the recorded speech includes video.

26. The system of claim 25 wherein the video is searched based upon the full text index.

27. The system of claim 14 wherein the recorded speech comprises a news program.

28. A computer readable medium containing program instructions for searching recorded speech in a database, the program instructions for:

(a) converting the recorded speech to text utilizing a voice recognition system, wherein breaks in the recorded speech are used to provide a plurality of time stamps;

(b) creating a full text index of the recorded speech utilizing an information extender, the full text index containing the plurality of time stamps which point to the occurrence of words in the recorded speech;

(c) searching the text by a full text server utilizing the full text index; and

(d) storing the searched text, the full text index and the recorded speech in the database, wherein specific contents of the recorded speech can be played back utilizing the full text index without listening to the complete recording.

29. The computer readable medium of claim 28 wherein the converting step (a) includes the step of (a1), storing the recorded speech in an index tool utilizing an audio extender.

30. The computer readable medium of claim 28 wherein the plurality of time stamps are produced during step (a).

31. The computer readable medium of claim 28 wherein the recorded speech is voicemail messages provided by an Internet-based voicemail system.

32. The computer readable medium of claim 31 wherein the voicemail messages are played back by accessing the voicemails, saying the keywords and listening to the phrases that contain the key words.

33. The computer readable medium of claim 28 wherein the recorded speech is voicemail messages provided by an Internet-based voicemail system.

34. The computer readable medium of claim 33 wherein a voicemail server is accessed via the Internet, search words and/or phrases of the voicemail message, observing a summary of the search results and listening to the relevant portions of the messages.

35. The computer readable medium of claim 28 wherein the recorded speech comprises a recorded meeting.

36. The computer readable medium of claim 35 wherein a recorded meeting is played back by listening to relevant portions of the recorded meeting, typing words and/or phrases that are being looked for, observing a textual summary of what was found and listening to the relevant contents based upon the textual summary.

37. The computer readable medium of claim 35 wherein the recorded meeting is played back by listening to relevant portions of the speech, saying the words and/or phrases that are being looked for and listening to the relevant contents.

38. The computer readable medium of claim 28 wherein the recorded speech includes video.

39. The computer readable medium of claim 38 wherein the video is searched based upon the full text index.