Multi-pass speech analytics

• US 7,752,043 B2
• Filed: 09/29/2006
• Issued: 07/06/2010
• Est. Priority Date: 09/29/2006
• Status: Active Grant

• Current Assignee: Verint Americas Incorporated (Verint Systems Incorporated)
• Original Assignee: Verint Americas Incorporated (Verint Systems Incorporated)
• Inventors: Watson, Joseph
• Primary Examiner(s): Han; Qi
• Application Number:
  US11/540,322
• Publication Number:
  US 20080082329A1
• Time in Patent Office: 1,376 Days
• Field of Search: 704/235, 704/231, 704/251, 704/246, 704/270
• US Class Current: 704/235
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems

Abstract

Included are embodiments for multi-pass analytics. At least one embodiment of a method includes receiving audio data associated with a communication, performing first tier speech to text analytics on the received audio data, and performing second tier speech to text analytics on the received audio.

Claims

1. A method for multi-pass analytics, comprising:

monitoring a communication for audio data;

receiving audio data associated with the communication;

performing first tier speech to text analytics on the received audio data, the first tier speech to text analytics including processing the received audio data in a first manner;

determining whether to perform second tier speech to text analytics;

in response to determining, performing second tier speech to text analytics on the received audio data, the second tier speech to text analytics including processing the received audio in a second manner;

forecasting an optimum level of service;

scheduling a staffing level to achieve the forecasted optimum level of service;

determining, from the scheduled staffing, at least one area for improvement; and

readjusting scheduling according to the at least one area for improvement.

2. The method of claim 1, further comprising determining recognition criteria associated with the audio data, the recognition criteria being configured to indicate whether to perform second tier speech to text analytics on the received audio.

3. The method of claim 2, wherein determining whether to perform second tier speech to text analytics includes determining, from the recognition criteria, whether to perform second tier speech to text analytics on the received audio.

4. The method of claim 1, wherein processing the audio data in a first manner includes performing a speech to text conversion of the audio data and processing the audio in a second manner includes analyzing the converted audio data.

5. The method of claim 4, wherein performing a speech to text conversion of the audio data includes utilizing at least one of the following:

phonetic speech to text conversion and Large Vocabulary Continuous Speech Recognition (LVCSR) speech to text conversion.

6. The method of claim 4, wherein analyzing the converted audio data includes at least one of the following:

identifying a speaker, verifying a speaker, detecting speaker emotion, and detecting speaker confidence.

7. The method of claim 1, wherein determining whether to perform second tier speech recognition analytics includes receiving indication from an analyst to perform second tier speech to text analytics.

8. The method of claim 1, wherein processing the audio data in a first manner includes performing at least one of the following:

automated call evaluation, automated call scoring, quality monitoring, quality assessment, script compliance, speech analytics used in conjunction with screen data, and fraud detection.

9. The method of claim 1, wherein processing the audio data in a second manner includes performing at least one of the following:

automated call evaluation, automated call scoring, quality monitoring, quality assessment, script compliance, speech analytics used in conjunction with screen data, and fraud detection.

10. The method of claim 1, wherein:

processing the audio data in a first manner includes performing a first speech to text conversion of the received audio data; and

processing the audio data in a second manner includes performing a second speech to text conversion of the received audio data, the second speech to text conversion of the received audio data being performed at a higher accuracy threshold than the first speech to text conversion.

11. A system for multi-pass analytics, comprising:

a receiving component configured as a recorder to receive audio data associated with a communication from a monitoring component;

a first tier speech to text analytics component executing on a first server configured to perform first tier speech to text analytics on the received audio data in a first manner;

a performance determining component configured to determine whether to perform second tier speech to text analytics;

a second tier speech to text analytics component executing on a second server configured to, in response to determining, perform second tier speech to text analytics on the received audio data in a second manner;

a forecasting component to determine an optimum level of service;

a scheduling component to schedule a staffing level to achieve the forecasted optimum level of service,

wherein it is determined, from the scheduled staffing, at least one area for improvement, and wherein the scheduling is readjusted according to the at least one area for improvement.

12. The system of claim 11, further comprising a criteria determining component configured to determine recognition criteria associated with the audio data, the recognition criteria being configured to indicate whether to perform second tier speech to text analytics on the received audio.

13. The system of claim 12, wherein the performance determining component is further configured to determine, from the recognition criteria, whether to perform second tier speech to text analytics on the received audio.

14. The system of claim 11, wherein performing speech to text analytics in a first manner includes performing a speech to text conversion of the audio data and performing speech to text analytics in a second manner includes analyzing the converted audio data.

15. The system of claim 14, wherein performing a speech to text conversion of the audio data includes utilizing at least one of the following:

a phonetic speech to text conversion component and a Large Vocabulary Continuous Speech Recognition (LVCSR) speech to text conversion component.

16. The system of claim 11, wherein analyzing the converted audio data includes at least one of the following:

identifying a speaker, verifying a speaker, detecting speaker emotion, and detecting speaker confidence.

17. The system of claim 11, wherein the performance determining component is further configured to receive indication from an analyst to perform second tier speech to text analytics.

18. The system of claim 11, wherein:

performing first tier speech to text analytics in a first manner includes performing a first speech to text conversion of the received audio data; and

performing first tier speech to text analytics in a second manner includes performing a second speech to text conversion of the received audio data, the second speech to text conversion of the received audio data being performed at a higher accuracy threshold than the first speech to text conversion.

19. A non-transitory computer readable storage medium storing containing computer executable instructions that when executed by a computing device perform a method for multi-pass analytics, comprising:

monitoring a communication for audio data;

receiving audio data associated with the communication;

performing first tier speech to text analytics on the received audio data in a first manner;

determining whether to perform second tier speech to text analytics;

in response to determining, performing second tier speech to text analytics on the received audio data in a second manner;

forecasting an optimum level of service;

scheduling a staffing level to achieve the forecasted optimum level of service;

determining, from the scheduled staffing, at least one area for improvement; and

readjusting scheduling according to the at least one area for improvement.

20. The non-transitory computer readable storage medium of claim 19, further comprising instructions for determining recognition criteria associated with the audio data, the recognition criteria being configured to indicate whether to perform second tier speech to text analytics on the received audio.

21. The non-transitory computer readable storage medium of claim 20, wherein the performance determining further configured to determine, from the recognition criteria, whether to perform second tier speech to text analytics on the received audio.

22. The non-transitory computer readable storage medium of claim 19, wherein performing speech to text analytics in a first manner includes performing a speech to text conversion of the audio data and performing speech to text analytics in a second manner includes analyzing the converted audio data.

23. The non-transitory computer readable storage medium of claim 22, wherein performing a speech to text conversion of the audio data includes utilizing at least one of the following:

phonetic speech to text conversion logic and Large Vocabulary Continuous Speech Recognition (LVCSR) speech to text conversion logic.

24. The non-transitory computer readable storage medium of claim 19, wherein the second tier speech to text analytics logic is further configured to perform at least one of the following:

identifying a speaker, verifying a speaker, detecting speaker emotion, and detecting speaker confidence.

25. The non-transitory computer readable storage medium of claim 19, wherein:

performing first tier speech to text analytics in a first manner includes performing a first speech to text conversion of the received audio data and analyzing the convened audio data; and

performing first tier speech to text analytics in a first manner includes performing a second speech to text conversion of the received audio data, the second speech to text conversion of the received audio data being performed at a higher accuracy threshold than the first speech to text conversion.