Multi-pass speech analytics

Abstract

Included are embodiments for multi-pass analytics. At least one embodiment of a method includes receiving audio data associated with first communications desired to be analyzed for an event, performing a phonetics analytics process on the audio data associated with the first communications to determine second communications associated with the event, and performing a large vocabulary continuous speech recognition (LVCSR) analytics process on audio data associated with the second communications to determine third communications associated with the event.

1 . A method of performing speech analytics, the method comprising:
 receiving audio data associated with first communications desired to be analyzed for an event;   performing a phonetics analytics process on the audio data associated with the first communications to determine second communications associated with the event; and   performing a large vocabulary continuous speech recognition (LVCSR) analytics process on audio data associated with the second communications to determine third communications associated with the event.   
     
     
         2 . The method of  claim 1  wherein performing the phonetics analytics process comprises processing the audio associated with the first communications to generate a phonetics index. 
     
     
         3 . The method of  claim 2  wherein performing the phonetics analytics process further comprises processing the phonetics index based on the event to determine the second communications associated with the event. 
     
     
         4 . The method of  claim 1  wherein performing the LVCSR analytics process comprises processing the audio data associated with the second communications to generate an LVCSR index. 
     
     
         5 . The method of  claim 4  wherein performing the LVCSR analytics process comprises processing the LVCSR index based on the event to determine the third communications associated with the event. 
     
     
         6 . The method of  claim 1  wherein the second communications comprise a subset of the first communications. 
     
     
         7 . The method of  claim 6  wherein the third communications comprises a subset of the second communications. 
     
     
         8 . The method of  claim 7  wherein the first communications comprise recorded calls. 
     
     
         9 . A system for performing speech analytics, the system comprising:
 an interface configured to receive audio data associated with first communications desired to be analyzed for an event; and   a processor coupled to the interface and configured to perform a phonetics analytics process on the audio data associated with the first communications to determine second communications associated with the event, and perform a large vocabulary continuous speech recognition (LVCSR) analytics process on audio data associated with the second communications to determine third communications associated with the event.   
     
     
         10 . The system of  claim 9  wherein the processor, to perform the phonetics analytics process, is configured to process the audio associated with the first communications to generate a phonetics index. 
     
     
         11 . The system of  claim 10  wherein the processor, to perform the phonetics analytics process, is further configured to process the phonetics index based on the event to determine the second communications associated with the event. 
     
     
         12 . The system of  claim 9  wherein the processor, to perform the LVCSR analytics process, is configured to process the audio data associated with the second communications to generate an LVCSR index. 
     
     
         13 . The system of  claim 12  wherein the processor, to perform the LVCSR analytics process, is further configured to process the LVCSR index based on the event to determine the third communications associated with the event. 
     
     
         14 . The system of  claim 9  wherein the second communications comprise a subset of the first communications. 
     
     
         15 . The system of  claim 14  wherein the third communications comprises a subset of the second communications. 
     
     
         16 . The method of  claim 15  wherein the first communications comprise recorded calls. 
     
     
         17 . A computer readable medium having stored thereon instructions that, when executed by a processor, direct the processor to:
 receive audio data associated with first communications desired to be analyzed for an event;   perform a phonetics analytics process on the audio data associated with the first communications to determine second communications associated with the event; and   perform a large vocabulary continuous speech recognition (LVCSR) analytics process on audio data associated with the second communications to determine third communications associated with the event.   
     
     
         18 . The computer readable medium of  claim 17  wherein the instructions, to perform the phonetics analytics process, direct the processor to process the audio associated with the first communications to generate a phonetics index. 
     
     
         19 . The computer readable medium of  claim 18  wherein the instructions, to perform the phonetics analytics process, further direct the processor to process the phonetics index based on the event to determine the second communications associated with the event. 
     
     
         20 . The computer readable medium of  claim 17  wherein the instructions, to perform the LVCSR analytics process, direct the processor to process the audio data associated with the second communications to generate an LVCSR index. 
     
     
         21 . The computer readable medium of  claim 20  wherein the instructions, to perform the LVCSR analytics process, further direct the processor to process the LVCSR index based on the event to determine the third communications associated with the event. 
     
     
         22 . The computer readable medium of  claim 17  wherein the second communications comprise a subset of the first communications. 
     
     
         23 . The computer readable medium of  claim 22  wherein the third communications comprises a subset of the second communications. 
     
     
         24 . The computer readable medium of  claim 23  wherein the first communications comprise recorded calls.