Methods and apparatus for audio data monitoring and evaluation using speech recognition

• US 7,076,427 B2
• Filed: 10/20/2003
• Issued: 07/11/2006
• Est. Priority Date: 10/18/2002
• Status: Expired due to Term

• Current Assignee: RingCentral Incorporated
• Original Assignee: SER Solutions Incorporated (Atos SE)
• Inventors: Scarano, Robert, Mark, Lawrence
• Primary Examiner(s): HARPER, V PAUL
• Application Number:
  US10/687,702
• Publication Number:
  US 20040117185A1
• Time in Patent Office: 995 Days
• Field of Search: 379/242, 379/265.02, 379/265.06, 704/231, 704/243, 704/245, 704/246, 704/251, 704/252, 704/253, 704/254, 704/270
• US Class Current: 704/270
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/2281 Call monitoring, e.g. for law enforcement purposes; Call tracing; Detection or prevention of malicious calls
• CPC Class Codes: H04M 3/5183 Call or contact centers with computer-telephony arrangements

Abstract

The present invention relates to audio data monitoring using speech recognition technology. In particular, the present invention uses business rules combined with unrestricted, natural speech recognition to monitor conversations in a customer interaction environment, literally transforming the spoken word to a retrievable data form. Implemented using the VorTecs Integration Platform (VIP), a flexible Computer Telephony Integration base, the present invention enhances quality monitoring by effectively evaluating conversations and initiating actionable events while observing for script adherence, compliance and/or order validation.

Claims

1. A method of analyzing audio data, comprising the step of:

processing an audio segment into a format suitable for rapid searching;

determining, in response to data associated with said audio segment, an appropriate set of business rules to apply to said audio segment; and

searching said audio segment in accordance with said appropriate set of business rules.

2. The method according to claim 1 further comprising a step of referencing said audio segment wherein said audio segment has been previously stored in an electronic media.

3. The method according to claim 1 further comprising a step of recording said audio segment.

4. The method according to claim 1 wherein said step of processing includes a step of processing said audio segment into a format suitable for rapid phonetic searching.

5. The method according to claim 1 wherein said step of processing includes a step of identifying symbols, corresponding to discrete portions of said audio segment.

6. The method according to claim 5 wherein said symbols represent respective phonemes of a set of phonemes characteristic of speech.

7. The method according to claim 1 wherein said step of searching includes the steps of:

attempting to find a match within said audio segment of a target phrase; and

in response, determining whether said target phrase is present within said audio segment at or above a specified confidence level.

8. The method according to claim 7 further comprising a step of triggering an event in response to said step of determining whether said target phrase is present within said audio segment.

9. The method according to claim 1 further comprising a step of triggering an event as a result of said searching step resulting in matching a given phrase at or above a specified confidence level.

10. The method according to claim 1 further comprising a step of triggering an event as a result of said searching step resulting in not finding a match for a given phrase at or above a specified confidence level.

11. The method according to claim 1 further comprising a step of incrementing a statistical parameter as a result of said searching step resulting in matching a given phrase at or above a specified confidence level.

12. The method according to claim 1 further comprising a step of incrementing a statistical parameter as a result of said searching step resulting in not finding a match for a given phrase at or above a specified confidence level.

13. The method according to claim 1 wherein said step of searching includes a step of searching said audio segment for a combination of a plurality of phrases.

14. The method according to claim 13 wherein said step of searching said audio segment for said combination of phrases includes a specified order of said phrases within said audio segment.

15. The method according to claim 14 further comprising the step of triggering an event in response to finding a match for said combination of phrases in said specified order in said audio segment.

16. The method according to claim 14 further comprising the step of triggering an event in response to not finding a match for said combination of phrases in said specified order in said audio segment.

17. The method according to claim 14 further comprising the step of incrementing a statistical value in response to finding a match for said combination of phrases in said specified order in said audio segment.

18. The method according to claim 14 further comprising the step of incrementing a statistical value in response to not finding a match for said combination of phrases in said specified order in said audio segment.

19. The method according to claim 13 wherein said step of searching said audio segment for said combination of phrases includes a specified temporal relationship of said phrases within said audio segment.

20. The method according to claim 19 wherein said temporal relationship comprises an occurrence of said phrases within a specified time period within said audio segment.

21. The method according to claim 1 wherein said step of searching includes a step of searching said audio segment for a target phrase occurrence within a specified time period within said audio segment.

22. The method according to claim 1 further comprising the steps of:

analyzing Computer Telephony Integration (CTI) data associated with said audio segment; and

providing an indication of satisfaction of a criteria in response to said steps of searching and analyzing.

23. The method according to claim 22 wherein said step of analyzing said CTI data includes a step of analyzing CTI data selected from the set consisting of (i) called number (dialed number identification service or “

DNIS”

) and (ii) calling number (Automatic Number Identification or “

ANT”

).

24. The method according to claim 1 further comprising a step of performing order validation.

25. The method according to claim 24 wherein said step of performing order validation includes the step of comparing a parameter of an order associated with said audio segment with a content of said audio segment resulting from said searching step.

26. The method according to claim 1 wherein said step of searching includes a step of searching for a target phrase, said method further comprising a step of performing order validation including determining whether an order associated with said audio segment is consistent with a result of said step of searching for said target phrase.

27. The method according to claim 26 further comprising a step of entering data for said order wherein said step of performing order validation includes validating whether said data is reflected within said audio segment.

28. The method according to claim 1 wherein said step of searching includes searching for a target utterance selected in response to said data related to said audio segment.

29. A method of processing audio data, comprising the step of:

importing call data;

selectively, responsive to said call data, analyzing an audio segment associated with said call data, said step of analyzing including

processing said audio segment into a format suitable for rapid searching;

determining, in response to said call data, an appropriate set of business rules to apply to said audio segment; and

searching said audio segment in accordance with said appropriate set of business rules.

30. The method according to claim 29 wherein said call data includes Computer Telephony Integration data selected from the group consisting of (i) called number (dialed number identification service or “

DNIS”

) and (ii) calling number (Automatic Number Identification or “

ANI”

).

31. The method according to claim 29 further comprising the steps of:

receiving call related event data associated with a telephone call, said call related event data related to said audio segment;

extracting said audio segment from said telephone call; and

correlating said data related to said audio segment to said audio segment.

32. The method according to claim 31 wherein said data related to said audio segment includes metadata.

33. The method according to claim 31 wherein said call related event data includes information selected from the group consisting of (i) time/day of call;

(ii) telephone number of a client party;

(iii) extension number of an agent; and

(iv) trunk identification.

34. The method according to claim 31 wherein said call related event data includes data selected from the group consisting of (i) dialed number identification service (DNIS);

(ii) Automatic Number Identification/Calling Line Identification (ANT/CLID);

(iii) collected digital; and

(iv) agent identification.

35. A system for analyzing audio data comprising:

an audio processor operable to process an audio segment into a format suitable for rapid searching;

logic responsive to data associated with said audio segment to determine an appropriate set of business rules to apply to said audio segment; and

a search engine operable to search said audio segment in accordance with said appropriate set of business rules.

36. The system according to claim 35 further comprising an electronic media having stored therein said audio segment and circuitry for retrieving said audio segment from said memory and providing said audio segment to said audio processor.

37. The system according to claim 35 further comprising an audio recorder operable to store said audio segment.

38. The system according to claim 35 wherein said audio processor is operable to process said audio segment into a format suitable for rapid phonetic searching and said search engine is operable to search said audio segment for phonetic information.

39. The system according to claim 35 wherein said search engine is further operable to identify symbols corresponding to discrete portions of said audio segment.

40. The system according to claim 39 wherein said symbols represent respective phonemes of a set of phonemes characteristic of speech.

41. The system according to claim 39 wherein said search engine is further operable to:

attempt to find a match within said audio segment of a target phrase; and

in response, determine whether said target phrase is present within said audio segment at or above a specified confidence level.

42. The system according to claim 41 further comprising logic operable to trigger an event in response to a presence or absence of said target phrase within said audio segment at or above said specified confidence level.

43. The system according to claim 35 further comprising logic operable to trigger an event in response to said search engine finding a target phrase within said audio segment at or above a specified confidence level.

44. The system according to claim 35 further comprising logic operable to trigger an event in response to said search engine not finding a target phrase within said audio segment at or above a specified confidence level.

45. The system according to claim 35 further logic operable to increment a statistical parameter as a result of said search engine finding a target phrase within said audio segment at or above a specified confidence level.

46. The system according to claim 35 further logic operable to increment a statistical parameter as a result of said search engine not finding a target phrase within said audio segment at or above a specified confidence level.

47. The system according to claim 35 wherein said search engine is further operable to search said audio segment for a combination of a plurality of phrases.

48. The system according to claim 47 wherein said search engine is further operable to search said audio segment for an occurrence of said combination of phrases in a specified order.

49. The system according to claim 48 further comprising logic operable to trigger an event in response to said search engine finding a match for said combination of phrases in said specified order in said audio segment.

50. The system according to claim 48 further comprising logic operable to trigger an event in response to said search engine not finding a match for said combination of phrases in said specified order in said audio segment.

51. The system according to claim 48 further comprising logic operable to increment a statistical value in response to said search engine finding a match for said combination of phrases in said specified order in said audio segment.

52. The system according to claim 48 further comprising logic operable to increment a statistical value in response to said search engine not finding a match for said combination of phrases in said specified order in said audio segment.

53. The system according to claim 48 wherein said search engine is further operable to search said audio segment for an occurrence of said combination of phrases in a specified temporal relationship within said audio segment.

54. The system according to claim 53 wherein said temporal relationship comprises an occurrence of said phrases within a specified time period within said audio segment.

55. The system according to claim 35 wherein said search engine is operable to search said audio segment for a target phrase occurrence within a specified time period within said audio segment.

56. The system according to claim 35 further comprising

logic operable to analyze Computer Telephony Integration (CTI) data associated with said audio segment and provide an indication of satisfaction of a criteria in response to said CTI data and an output from said search engine.

57. The system according to claim 56 wherein said logic operable to analyze said CTI data is responsive to CTI data selected from the set consisting of (i) called number (dialed number identification service or “

DNIS”

) and (ii) calling number (Automatic Number Identification or “

ANI”

).

58. The system according to claim 35 further comprising logic operable to perform order validation.

59. The system according to claim 58 wherein said logic operable to perform order validation is operable to compare a parameter of an order associated with said audio segment with a content of said audio segment identified by said search engine.

60. The system according to claim 35 wherein said search engine is further operable to search for a target phrase, said system further comprising logic operable to perform order validation including determining whether an order associated with said audio segment is consistent with a result of said search engine searching for said target phrase.

61. The system according to claim 60 further comprising a terminal operable for the entry of data for said order wherein said logic operable to perform said order validation is operable to validate whether said data is reflected within said audio segment.

62. The system according to claim 35 further comprising:

circuitry for receiving a call related event data associated with a telephone call, said call related event data related to said audio segment;

logic for extracting said audio segment from said telephone call; and

logic for correlating said data to said audio segment.

63. The system according to claim 62 wherein said data related to said audio segment includes metadata including call data augmenting an information content directly extractable from said audio segment.

64. The system according to claim 35 further operable for:

receiving a call related event data associated with a telephone call, said call related event data related to said audio segment;

extracting said audio segment from said telephone call; and

correlating said data to said audio segment.

65. The system according to claim 64 wherein said call related event data includes information selected from the group consisting of (i) time/day of call;

(ii) telephone number of a client party;

(iii) extension number of an agent; and

(iv) trunk identification.

66. The system according to claim 64 wherein said call related event data includes data selected from the group consisting of (i) dialed number identification service (DNIS);

(ii) Automatic Number Identification/Calling Line Identification (ANI/CLID);

(iii) collected digital; and

(iv) agent identification.

67. The system according to claim 35 wherein said step of search engine is further operable to search for a target utterance selected in response to said data related to said audio segment.

68. A system of processing audio data comprising:

telephone equipment connected to receive call data;

an audio processor responsive to said call data for selectively analyzing an audio segment associated with said call data, said audio processor operable to

process said audio segment into a format suitable for rapid searching;

determine, in response to said call data, an appropriate set of business rules to apply to said audio segment; and

search said audio segment in accordance with said appropriate set of business rules.

69. The system according to claim 68 wherein said call data includes Computer Telephony Integration data selected from the group consisting of (i) called number (dialed number identification service or “

DNIS”

) and (ii) calling number (Automatic Number Identification or “

ANI”

).

70. A method for monitoring audio data, comprising:

recording an audio segment;

setting business rules, in response to metadata associated with said audio segment, for searching for spoken words or phrases in said audio segment using speech recognition technology;

searching said audio segment in accordance with said business rules; and

providing a report based on said search.