CORRELATED CALL ANALYSIS

• US 20100161315A1
• Filed: 12/24/2008
• Published: 06/24/2010
• Est. Priority Date: 12/24/2008
• Status: Active Grant

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: LJOLJE, Andrej, KIM, Yeon-Jun, MELAMED, I. Dan, RENGER, Bernard S., SMITH, David J.
• Granted Patent: US 8,756,065 B2
• Time in Patent Office: Days
• US Class Current: 704/9
• CPC Class Codes: G06F 40/216 using statistical methods
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques

Abstract

A method of correlating received communication data with operational communication characteristics is provided. The method includes receiving audible input from a source in a communication over a communications network, recording the received audible input, and transcribing the recorded audible input into a transcript. The method further includes outputting the transcript, specifying features of the transcript to be analyzed, specifying and recording operational communication characteristics particular to the communication, analyzing the transcript for the specified features to identify patterns associated with the audible input, computing statistical correlations of the identified patterns with the operational communication characteristics, and outputting results of the computed statistical correlations on a user interface.

Claims

1. A method of correlating received communication data with operational communication characteristics, comprising:

receiving audible input from a source in a communication over a communications network, recording the received audible input, and transcribing the recorded audible input into a transcript;

outputting the transcript;

specifying features of the transcript to be analyzed;

specifying and recording the operational communication characteristics particular to the communication;

analyzing the transcript for the specified features to identify patterns associated with the audible input; and

computing statistical correlations of the identified patterns with the operational communication characteristics.

2. The method of claim 1, wherein analyzing the transcript further comprises:

identifying repeated audible patterns corresponding to the audible input;

identifying unique audible patterns corresponding to the audible input;

storing identified repeated and unique audible patterns corresponding to the audible input; and

outputting stored repeated and unique audible patterns.

3. The method of claim 1, wherein the specified features include at least one speech recognition parameter.

4. The method of claim 3, wherein the at least one speech recognition parameter comprise at least one of an acoustic, a prosodic, a lexical, a phonetic, a morphological, a syntactic, a semantic, a discourse, a dialog, and a speaker identification parameter.

5. The method of claim 1, wherein the operational communication characteristics include information about at least one of a time, a date, an area code, a call duration, a geographic location, a network designation, an agent identifier, and a product identifier.

6. The method of claim 1, further comprising:

displaying the outputted results to determine at least one of customer satisfaction, call duration, and revenue generation within the communications network.

7. The method of claim 1, further comprising:

transferring the received communication to a post-communication information gathering processor;

requesting post-communication information input from the received communication;

detecting results of the requested post-communication information input;

outputting results of the requested post-communication information input; and

computing statistical correlations of the identified patterns, the operational communication characteristics, and the results of the requested post-communication information input.

8. The method of claim 7, further comprising:

receiving post-communication information audible input from the received communication;

recording the post-communication information audible input from the received communication;

transcribing the recorded post-communication information audible input into a post-communication information transcript;

outputting the post-communication information transcript;

specifying features of the post-communication information transcript to be analyzed;

analyzing the post-communication information transcript for the specified features to identify patterns associated with the post-communication information audible input; and

computing statistical correlations of the identified patterns of the post-communication information transcript with the operational communication characteristics.

9. The method of claim 8, further comprising:

identifying repeated audible patterns corresponding to the post-communication information audible input;

identifying unique audible patterns corresponding to the post-communication information audible input;

storing identified repeated and unique audible patterns corresponding to the post-communication information audible input; and

outputting stored repeated and unique audible patterns corresponding to the post-communication information audible input.

10. The method of claim 1, wherein the identified patterns comprise at least one of:

a number of words;

numbers;

repeated words;

words associated with emotional indicators;

repeated prefixes;

repeated suffixes;

negative words; and

positive words.

11. The method of claim 1, further comprising:

collecting a plurality of computed statistical correlations from a plurality of received communications to identify trends within the communications network for determining at least one of customer satisfaction, call duration, and revenue within the communications network.

12. The method of claim 11, further comprising:

activating alerts configured to identify certain correlation trends; and

alerting a user within the communications network when the certain identified trends are produced.

13. A system for correlating received communication data with operational communication characteristics, comprising:

a recorder that receives and records audible input from a received communication over a communications network;

a speech-recognition analyzer that communicates with the recorder to transcribe the recorded audible input into a transcript;

a speech-recognition database that includes speech-recognition features;

a transcript analyzer that communicates with the speech-recognition analyzer and the speech-recognition database to analyze the transcript based on the speech-recognition features, and identifies patterns associated with the audible input;

an operational communication characteristics processor that collects and records the operational communication characteristics;

a correlator that communicates with the transcript analyzer and the operational communication characteristics processor to compute statistical correlations of the identified patterns associated with the audible input and the recorded operational communication characteristics; and

an outputter to display results of the computed statistical correlations.

14. The system of claim 13, wherein the transcript analyzer is configured to identify unique and repeated audible patterns corresponding to the audible input, to store the identified unique and repeated audible patterns, and to output the stored unique and repeated audible patterns to the correlator.

15. The system of claim 13, wherein the speech-recognition features include at least one speech recognition parameter.

16. The system of claim 15, wherein the at least one speech recognition parameter comprise at least one of an acoustic, a prosodic, a lexical, a phonetic, a morphological, a syntactic, a semantic, a discourse, a dialog, and a speaker identification parameter.

17. The system of claim 13, wherein the operational communication characteristics include information about at least one of a time, a date, an area code, a call duration, a geographic location, a network designation, an agent identifier, and a product identifier.

18. The system of claim 13, wherein the outputter is configured to display results based on at least one of customer satisfaction, call duration, and revenue within the communications network.

19. A tangible computer readable storage medium comprising a computer program for processing input variables in a system for correlating received communication data with operational communication characteristics, the tangible computer readable storage medium comprising:

a receiving code segment, recorded on the tangible computer readable storage medium, executable to receive, record and transcribe received communication audible input into a transcript;

an analyzing code segment, recorded on the tangible computer readable storage medium, executable to analyze the transcript based on a speech-recognition database that includes speech-recognition features;

a collecting code segment, recorded on the tangible computer readable storage medium, executable to collect and record operational communication characteristics;

a correlating code segment, recorded on the tangible computer readable storage medium, executable to correlate the analyzed transcript with the recorded operational communication characteristics; and

an output code segment, recorded on the tangible computer readable storage medium, executable to display results of the correlations.

20. The computer program of claim 19, wherein the analyzing code segment further comprises:

an identifying code segment, recorded on the tangible computer readable storage medium, executable to identify unique and repeated audible patterns based on the audible input; and

a storing code segment, recorded on the tangible computer readable storage medium, executable to store and to output the unique and repeated audible patterns for correlation with the recorded operational communication characteristics.