Functions and associated communication capabilities for a speech analytics component to support agent compliance in a call center

• US 9,602,665 B1
• Filed: 08/16/2013
• Issued: 03/21/2017
• Est. Priority Date: 07/24/2013
• Status: Active Grant

• Current Assignee: Alvaria Cayman
• Original Assignee: Noble Systems Corporation
• Inventors: Koster, Karl H.
• Primary Examiner(s): AL AUBAIDI, RASHA S
• Application Number:
  US13/968,692
• Time in Patent Office: 1,313 Days
• Field of Search: 37926501-26502, 379/266.07, 379/88.01, 379/265.07
• US Class Current: 1/1
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: H04M 2201/40 using speech recognition speech recognition per se G10L15/00
• CPC Class Codes: H04M 2203/401 Performance feedback
• CPC Class Codes: H04M 2203/552 Call annotations
• CPC Class Codes: H04M 3/42221 Conversation recording systems at the subscriber's set H04M1/656
• CPC Class Codes: H04M 3/5175 Call or contact centers supervision arrangements

Abstract

A command set and associated capabilities are defined for a speech analytics component (“SAC”) in a call center. The speech analytics component is used to monitor speech of an agent and a remote party on a call for the purpose of determining whether the agent is complying with various call center policies or applicable regulations. The command set allows the call handler to instruct the SAC to load a keyword set into memory for future use, apply the keyword set to an indicated call and for an indicated party, to remove use of the keyword set, and to release the keyword set from memory. This allows the SAC to monitor speech as required for a particular context, which is known to the call handler, but not necessarily by the SAC. In this way, efficient use of the SAC processing resource is obtained for various call contexts.

Claims

1. A speech analytics component in a call center comprising:

a first interface for receiving a voice call from a call handler identified by a call reference value, the voice call comprising a first data stream associated with an agent and a second data stream associated with a remote party;

a second interface for receiving commands from the call handler;

a memory configured to store a plurality of keyword sets; and

a microprocessor configured to;

receive the voice call at the first interface associated with the call reference value,

receive a first command at the second interface identifying a keyword set from among the plurality of keyword sets, the call reference value, and a party identifier associated with a party comprising at least one of the agent and the remote party, wherein the first command indicates application of the keyword set for analyzing speech of the party on the voice call,

analyze the speech from the party using the keyword set identified from among the plurality of keyword sets to determine whether the party speaks a keyword in the keyword set, and

provide an indication over the second interface to the call handler indicating detection of the keyword in the keyword set in the speech from the party.

2. The speech analytics component of claim 1, wherein the first interface and the second interface comprise IP based protocols for conveying voice data of the voice call and control data respectively.

3. The speech analytics component of claim 1, wherein the microprocessor is further configured to:

receive a second command at the second interface from the call handler terminating application of the keyword set for analyzing the speech from the party.

4. The speech analytics component of claim 3, wherein the microprocessor is further configured to:

receive a third command at the second interface from the call handler after receiving the second command, the third command instructing a second keyword set to be applied for analyzing speech of another party comprising at least one of the agent and the remote party, wherein the another party is different from the party.

5. The speech analytics component of claim 1, wherein the microprocessor is further configured to:

receive a second command at the second interface from the call handler indicating that the keyword set is no longer required to be stored in the memory.

6. The speech analytics component of claim 1, wherein the microprocessor is further configured to:

load the keyword set into the memory in response to receiving a command from the call handler identifying the keyword set to be loaded.

7. The speech analytics component of claim 1, wherein the indication provided to the call handler identifies the keyword in the keyword set detected in the speech of the party.

8. The speech analytics component of claim 1, wherein the keyword set comprises a plurality of keywords.

9. A method for processing speech from a call in a call center comprising:

establishing a call leg from a call handler to a speech analytics component in the call center, wherein the call leg is identified by a call reference value, wherein the call reference value is associated with the call, and the call is between an agent and a remote party;

providing a first message from the call handler to the speech analytics component identifying the call reference value, a party identifier of a party on the call, and a first keyword set from among a plurality of keyword sets to be used by the speech analytics component for analyzing speech of the party, wherein the party comprises at least one of the agent and the remote party;

analyzing the speech of the party using the first keyword set by the speech analytics component to detect a keyword from the first keyword set in the speech of the party identified by the party identifier; and

transmitting an indication from the speech analytics component to the call handler indicating detection of the keyword in the speech of the party and the call reference value.

10. The method of claim 9, wherein the first message and the call are conveyed from the call handler to the speech analytics component using an Internet based protocol on a communication facility.

11. The method of claim 9 further comprising:

providing a second message from the call handler to the speech analytics component after the indication is transmitted from the speech analytics component, the second message instructing the speech analytics component to terminate application of the keyword set in analyzing the speech from the party.

12. The method of claim 11, further comprising:

providing a third message from the call handler to the speech analytics component after the second message is received from the call handler, the third message instructing the speech analytics component to apply a second keyword set from among the plurality of keyword sets to speech of another party comprising at least one of the agent and the remote party, wherein the another party is different from the party.

13. The method of claim 12, further comprising:

transmitting a second indication from the speech analytics component to the call handler indicating the call reference value and detection of a second keyword from the second keyword set in the speech of the another party; and

providing a fourth message from the call handler to the speech analytics component indicating the second keyword set is no longer to be applied to the speech of the another party.

14. The method of claim 11, further comprising:

providing a third message from the call handler to the speech analytics component after the second message is received from the call handler, the third message indicating that the keyword set is no longer required to be stored in a memory of the speech analytics component.

15. A computer-readable, non-transitory medium storing instructions that when executed by a processor cause the processor to:

store a call reference value associated with a first received call from a call handler at a first interface, wherein the first received call is between an agent and a remote party;

process a first command received identifying a keyword set from among a plurality of keyword sets, the call reference value, and a party indicator identifying one of the agent or the remote party, wherein the first command indicates application of the keyword set for detecting a keyword of the keyword set in speech by a party on the first received call;

determine the keyword of the keyword set was spoken by the party; and

transmit an indication over a second interface to the call handler indicating detection of the keyword in the keyword set in the speech from the party, the indication further comprising the party indicator and the call reference value.

16. The computer-readable, non-transitory medium of claim 15, wherein the instructions further cause the processor to:

process a second command comprising the call reference value, wherein the second command indicates terminating application of the keyword set from the speech of the party; and

in response to processing the second command, terminate application of the keyword set from the speech of the party.

17. The computer-readable, non-transitory medium of claim 16, wherein the instructions further cause the processor to:

process a third command identifying a second keyword set, the call reference value, and a another party identifier comprising one of the agent or the remote party, wherein the third command indicates application of the second keyword set for detecting a second keyword of the second keyword set in speech by another party on the first received call, and wherein the another party is different from the party;

determine the second keyword of the second keyword set was spoken by the another party; and

generate a second indication over the second interface to the call handler indicating detection of the second keyword in the second keyword set in the speech from the another party, the indication further comprising another party indicator and the call reference value.

18. The computer-readable, non-transitory medium of claim 17,

wherein the instructions further cause the processor to:

process a fourth command, wherein the fourth command indicates terminating application of the second keyword set from the speech of the another party; and

in response to processing the fourth command, terminate application of the second keyword set from the speech of the another party.

19. The computer-readable, non-transitory medium of claim 15, wherein the instructions further cause the processor to:

process a second command received prior to the first command, wherein the second command indicates retrieval of the keyword set from a memory store.

20. The computer-readable, non-transitory medium of claim 15, wherein the instructions further cause the processor to:

include the keyword detected from the keyword set in the indication to the call handler.