Systems and methods for analyzing communication sessions using fragments

• US 7,881,216 B2
• Filed: 09/29/2006
• Issued: 02/01/2011
• Est. Priority Date: 09/29/2006
• Status: Active Grant

• Current Assignee: Verint Americas Incorporated (Verint Systems Incorporated)
• Original Assignee: Verint Systems Incorporated
• Inventors: Blair, Christopher D.
• Primary Examiner(s): Nguyen; Steven H
• Assistant Examiner(s): Skripnikov; Alex
• Application Number:
  US11/540,353
• Publication Number:
  US 20080080385A1
• Time in Patent Office: 1,586 Days
• Field of Search: None
• US Class Current: 370/252
• CPC Class Codes: H04M 3/2227 Quality of service monitoring
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements

Abstract

Systems and methods for analyzing communication sessions using fragments are provided. In this regard, a representative method includes: delineating fragments of an audio component of a communication session, each of the fragments being attributable to a party of the communication session and representing a contiguous period of time during which that party was speaking; and automatically assessing quality of at least some of the fragments such that a quality assessment of the communication session is determined.

Claims

1. A method for analyzing communication sessions using fragments comprising:

delineating fragments of an audio component of a communication session by a communication analyzer, each of the fragments being attributable to a party of the communication session and representing a contiguous period of time during which that party was speaking; and

automatically assessing quality of at least some of the fragments to determine a quality assessment of the communication session by the communication analyzer, wherein automatically assessing a quality of at least some of the fragments comprises assigning a quality score to each of the at least some of the fragments and automatically assessing further comprises manually assessing the quality of at least some of the fragments and using the quality assessments obtained manually as training inputs for machine learning algorithms for automatically assessing the quality of other fragments, and

wherein the quality assessment is measured in accordance with a set of criteria that define the interaction between each party to the communication session.

2. The method of claim 1, wherein automatically assessing comprises analyzing a sequence of the fragments to determine which party was speaking and for how long.

3. The method of claim 1, wherein automatically assessing comprises defining rules and analyzing the fragments for characteristics embodied by the rules.

4. The method of claim 3, wherein the rules indicate that a quality assessment is to be lowered based on a determination that a party to a communication session is a contact center agent, and that the agent interrupted, by speaking, another party of the communication session that was speaking.

5. The method of claim 3, wherein the rules indicate that a quality assessment is to be lowered based on a determination that a party to a communication session is a contact center agent, and that the agent spoke for a duration exceeding a predetermined time limit without another party to the communication session speaking.

6. The method of claim 3, wherein the rules indicate that a quality assessment is to be lowered based on a determination that a party to a communication session is a contact center agent, and that the agent spoke at a volume level that at least one of:

not less than a high volume threshold and not higher than a low volume threshold.

7. The method of claim 1, wherein automatically assessing comprises, with respect to the fragments analyzed, weighting the quality score associated with the fragments based, at least in part, on a time that the respective fragments occurred during the communication session.

8. The method of claim 1, wherein automatically assessing comprises performing script adherence analysis.

9. The method of claim 1, wherein automatically assessing comprises evaluating the communication session for fraud.

10. The method of claim 1, wherein at least a portion of the communication session is conducted using Internet Protocol packets.

11. The method of claim 1, further comprising recording the communication session.

12. The method of claim 1, wherein:

one of the parties to the communication session is a contact center agent; and

the method further comprises altering a work schedule of the agent based, at least in part, on the quality assessment of the communication session.

13. The method of claim 1, further comprising appending information to the fragments.

14. A system for analyzing communications using fragments comprising:

a communication analyzer operative to;

delineate fragments of an audio component of a communication session, each of the fragments being attributable to a party of the communication session and representing a contiguous period of time during which that party was speaking; and

automatically assess quality of at least some of the fragments to determine a quality assessment of the communication session, wherein automatically assessing a quality of at least some of the fragments comprises assigning a quality score to each of the at least some of the fragments and automatically assessing further comprises manually assessing the quality of at least some of the fragments and using the quality assessments obtained manually as training inputs for machine learning algorithms for automatically assessing the quality of other fragments, and

wherein the quality assessment is measured in accordance with a set of criteria that define the interaction between each party to the communication session.

15. The system of claim 14, wherein the system comprises a speech recognition engine operative to generate a transcript of at least a portion of the communication session.

16. The system of claim 14, wherein the system comprises a phonetic analyzer operative to generate a phoneme sequence of at least a portion of the communication session.

17. The system of claim 14, wherein the system comprises a vox detection analyzer operative to provide amplitude information corresponding to volume levels that the audio component exhibited during the communication session, the volume levels being used by the communication analyzer to determine locations at for defining fragments.

18. A non-transitory computer readable medium having a computer program stored thereon for performing the computer executable method of:

delineating fragments of an audio component of a communication session, each of the fragments being attributable to a party of the communication session and representing a contiguous period of time during which that party was speaking; and

automatically assessing quality of at least some of the fragments to determine a quality assessment of the communication session, wherein automatically assessing a quality of at least some of the fragments comprises assigning a quality score to each of the at least some of the fragments and automatically assessing further comprises manually assessing the quality of at least some of the fragments and using the quality assessments obtained manually as training inputs for machine learning algorithms for automatically assessing the quality of other fragments, and

wherein the quality assessment is measured in accordance with a set of criteria that define the interaction between each party to the communication session.

19. The non-transitory computer readable medium of claim 18, wherein automatically assessing comprises analyzing a sequence of the fragments to determine which party was speaking and for how long.

20. The non-transitory computer readable medium of claim 18, wherein automatically assessing comprises defining rules and analyzing the fragments for characteristics embodied by the rules.