“At least” operator for combining audio search hits

• US 9,275,139 B2
• Filed: 09/27/2012
• Issued: 03/01/2016
• Est. Priority Date: 03/30/2012
• Status: Active Grant

• Current Assignee: Aurix Ltd. (Avaya Incorporated)
• Original Assignee: Aurix Ltd. (Avaya Incorporated)
• Inventors: Ponting, Keith Michael
• Primary Examiner(s): BAKER, MATTHEW H
• Application Number:
  US13/629,589
• Publication Number:
  US 20130262124A1
• Time in Patent Office: 1,251 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/632 Query formulation
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G10L 15/063 Training
• CPC Class Codes: G10L 15/18 using natural language modelling

Abstract

System and method to search audio data, including: receiving audio data representing speech; receiving a search query related to the audio data; compiling, by use of a processor, the search query into a hierarchy of scored speech recognition sub-searches; searching, by use of a processor, the audio data for speech identified by one or more of the sub-searches to produce hits; and combining, by use of a processor, the hits by use of at least one combination function to provide a composite search score of the audio data. The combination function may include an at-least-M-of-N function that produces a high score when at least M of N function inputs exceed a predetermined threshold value. The composite search score employ a soft time window such as a spline function.

Claims

1. A method to search audio data, comprising:

receiving audio data representing speech;

receiving a search query related to the audio data;

compiling, by use of a processor, the search query into a hierarchy of scored speech recognition sub-searches;

searching, by use of an audio search engine operating in an audio domain, the audio data for speech identified by one or more of the sub-searches to produce hits;

scoring the speech recognition by use of a soft time window, wherein the soft time window comprises a window definable by a spline function; and

combining, by use of a processor, the hits by use of at least one combination function to provide a composite search score of the audio data,

wherein at least one of the combination functions comprises an at-least-M-of-N function, wherein the at-least-M-of-N function produces a high score when at least M members found within a set having membership size of N exceed a predetermined threshold value, wherein N is greater than or equal to M.

2. The method of claim 1, wherein the high score is produced by applying a transform function to at least one of N function inputs.

3. The method of claim 1 wherein the at least-M-of-N function comprises an ALL-function, wherein the ALL-function produces a high score when all inputs to the combination function exceed a predetermined threshold value within a predetermined time window.

4. The method of claim 1 wherein a hit comprises an identifier that indicates what has been matched, a time interval over which the match is declared and a relevance score.

5. The method of claim 1, further comprising thresholding search results obtained from the step of searching the audio data for speech identified by one or more of the sub-searches.

6. The method of claim 1 wherein the audio data comprises content-dependent metadata and content-independent metadata.

7. The method of claim 6 wherein the content-dependent metadata comprises:

an identification of a predetermined speaker;

a time interval during which the speaker was speaking; and

a quality score of the identification of the predetermined speaker.

8. The method of claim 1 wherein the composite search score comprises an attribute with an associated score and or more associated time intervals.

9. The method of claim 1 wherein the composite search score may be determined in accordance with the following relationship:

10. The method of claim 9 wherein the sensitivity vectors are obtained from user feedback that indicates a correctness of the composite search score.

11. The method of claim 10, further comprising self-learning of the sensitivity vectors based upon the user feedback.

20. The method of claim 8, wherein the at-least-M-of-N function comprises an ALL-function, wherein the ALL-function produces a high score when all inputs to the combination function exceed a predetermined threshold value within a predetermined time window.

12. A system to search audio data, comprising:

a receiver configured to receive audio data representing speech;

a receiver configured to receive a search query related to the audio data;

a processor configured to compile the search query into a hierarchy of scored speech recognition sub-searches;

an audio search engine operating in an audio domain configured to search the audio data for speech identified by one or more of the sub-searches to produce hits; and

a processor configured to combine the hits by use of at least one combination function to provide a composite search score of the audio data,

wherein at least one of the combination functions comprises an at-least-M-of-N function, wherein the at-least-M-of-N function produces a high score when at least M of N function inputs exceed a predetermined threshold value,

wherein the composite search score may be determined in accordance with the following relationship;

13. The system of claim 12, wherein the high score is produced by applying a transform function to at least one of N function inputs.

14. The system of claim 12 wherein the audio data comprises content-dependent metadata and content-independent metadata.

15. The system of claim 12 wherein the composite search score comprises an attribute with an associated score and or more associated time intervals.

16. The system of claim 12, wherein the sensitivity vectors are obtained from user feedback that indicates a correctness of the composite search score.

17. The system of claim 16 further comprising a self-learning module to process the sensitivity vectors based upon the user feedback.

18. A method to search audio data, comprising:

receiving audio data representing speech;

receiving a search query related to the audio data;

compiling, by use of a processor, the search query into a hierarchy of scored speech recognition sub-searches;

searching, by use of an audio search engine operating in an audio domain, the audio data for speech identified by one or more of the sub-searches to produce hits; and

combining, by use of a processor, the hits by use of at least one combination function to provide a composite search score of the audio data,

wherein at least one of the combination functions comprises an at-least-M-of-N function, wherein the at-least-M-of-N function produces a high score when at least M members found within a set having membership size of N exceed a predetermined threshold value, wherein N is greater than or equal to M,

wherein the composite search score may be determined in accordance with the following relationship;

19. The method of claim 18, wherein the high score is produced by applying a transform function to at least one of N function inputs.

21. The method of claim 18, wherein a hit comprises an identifier that indicates what has been matched, a time interval over which the match is declared and a relevance score.

22. The method of claim 18, further comprising thresholding search results obtained from the step of searching the audio data for speech identified by one or more of the sub-searches.

23. The method of claim 18, wherein the audio data comprises content-dependent metadata and content-independent metadata.

24. The method of claim 23, wherein the content-dependent metadata comprises:

an identification of a predetermined speaker;

a time interval during which the speaker was speaking; and

a quality score of the identification of the predetermined speaker.

25. The method of claim 18, wherein the composite search score comprises an attribute with an associated score and or more associated time intervals.

26. The method of claim 18, wherein the sensitivity vectors are obtained from user feedback that indicates a correctness of the composite search score.

27. The method of claim 26, further comprising self-learning of the sensitivity vectors based upon the user feedback.

28. The method of claim 18, further comprising scoring the speech recognition by use of a soft time window.

29. The method of claim 28, wherein the soft time window comprises a window definable by a spline function.