Management system for using speech analytics to enhance agent compliance for debt collection calls
View Patent ↗A call center compliance system is used to ensure that agents adhere to call center policies and applicable regulations when handling calls. A call handler processing a call between an agent and a remote party bridges on a speech analytics component. A keyword set is used by the speech analytics component to analyze the remote party's or the agent's speech. If a keyword is detected, an event notification is provided from the speech analytics component to the call handler. A logic script executing in the call handler defines what actions are to occur in response to receiving the event notification. The call handler can provide information to the agent, record a violation, or await further event notifications. In one embodiment, detection of a first keyword in a first keyword set spoken by a remote party causes the agent's speech to be monitored using a second keyword set.
1. A method for responding to a remote party in communication with an agent in a call center, comprising:
recording a response script in a voice of the agent;
processing a call by a call handler, the call involving a first call leg to the remote party, a second call leg to the agent, and a third call leg to a speech analytics component;
processing speech from the remote party by the speech analytics component to detect a presence of a first keyword spoken by the remote party;
providing an event notification from the speech analytics component to the call handler signifying detection of the first keyword in the speech from the remote party; and
in response to receiving the event notification from the speech analytics component, presenting information on a computer screen for the agent to command the response script to be played to the remote party.
2. The method of claim 1 , wherein
the call to the remote party is made for purposes of attempting to collect a debt owed by the remote party, and
the first keyword detected from the remote party by the speech analytic component comprises one from a group of “bankrupt,” “bankruptcy,” and “chapter 7.”
3. The method of claim 1 , further comprising:
receiving a command from the agent to play the response script to the remote party; and
playing the response script to the remote party in response to receiving the command.
4. The method of claim 1 , further comprising:
presenting further information on the computer screen of the agent by the call handler, the further information associated with a preferred response to be provided by the agent after the remote party informs the agent of a bankruptcy filing by the remote party.
5. The method of claim 4 , wherein the agent does not command the call handler to play the response script to the remote party, and the method further comprises:
processing speech from the agent by the speech analytics component to ascertain a presence of a second keyword spoken by the agent, wherein the second keyword comprises one from a group of “attorney,” “lawyer,” “law firm,” and “case number.”
6. The method of claim 4 , wherein the further information comprises a text-based script configured to be read by the agent to the remote party.
7. The method of claim 6 , wherein
a timer is started after receiving the event notification from the speech analytics component indicating detection of the first keyword, and
the timer is stopped after receiving a command from the agent to play the response script to the remote party.
8. A call center system comprising:
a call handler configured to process a call between a remote party and an agent, the call handler comprising a first processor configured to
receive an event notification signifying detection of a first keyword spoken by the remote party, and
present information in response to receiving the event notification on a computer screen used by the agent for commanding a pre-recorded response in a voice of the agent to be played to the remote party; and
a speech analytics component comprising a second processor configured to
analyze speech from the remote party to detect a presence of the first keyword, and
provide the event notification to the call handler signifying detection of the first keyword spoken by the remote party.
9. The call center system of claim 8 , wherein the first keyword comprises one from a group of “bankrupt,” “bankruptcy,” and “chapter 7.”
10. The call center system of claim 8 , wherein the first processor is further configured to:
present further information on the computer screen used by the agent, the further information associated with a preferred response to be provided by the agent after detecting the presence of the first keyword spoken by the remote party.
11. The call center system of claim 10 , wherein the further information presented on the computer screen is configured as a text script to be read by the agent.
12. The call center system of claim 8 , wherein the agent does not command the call handler to play the pre-recorded response to the remote party, and the second processor is further configured to:
process speech from the agent to detect a presence of a second keyword, wherein the second keyword comprises one from a group of “lawyer,” “attorney,” and “law firm.”
13. The call center system of claim 8 , wherein the first processor is further configured to start a timer after receipt of the event notification and the timer is stopped upon receiving a command from the agent to play the pre-recorded response to the remote party.
14. The call center system of claim 8 , wherein the first processor is further configured to:
receive a command from the agent to play the pre-recorded response to the remote party; and
play the pre-recorded response to the remote party in response to receiving the command.
15. A computer-readable non-transitory medium storing instructions that when executed by a processor comprising an integrated circuit microprocessor, cause the processor to:
process a call involving a remote party and an agent in a call center, wherein a call leg of the call is established to a speech analytics component;
receive an event notification from the speech analytics component signifying detection of a first keyword spoken by the remote party during the call; and
in response to receiving the event notification, present an information on a computer screen used by the agent for commanding a pre-recorded response script in a voice of the agent to be played to the remote party.
16. The computer-readable non-transitory medium of claim 15 , wherein the instructions further cause the processor to:
receive a command from the agent to play the pre-recorded response script to the remote party; and
play the pre-recorded response script to the remote party in response to receiving the command.
17. The computer-readable non-transitory medium of claim 15 , wherein the call involves collection of a debt owed by the remote party, and the event notification received from the speech analytics component signifies detection of the first keyword comprising one from a group of “bankrupt,” “bankruptcy,” and “chapter 7.”
18. The computer-readable non-transitory medium of claim 17 , wherein the instructions further cause the processor to:
receive a second event notification from the speech analytics component signifying detection of the agent speaking of a second keyword from a keyword set comprising “lawyer,” “attorney,” and “law firm.”
19. The computer-readable non-transitory medium of claim 15 , wherein the instructions further cause the processor to:
initiate a timer after receipt of the event notification; and
terminate the timer after receipt of a command from the remote party to play the pre-recorded response script or after receipt of a second event notification from the speech analytics component signifying detection of a second keyword from a group comprising “lawyer,” “attorney” and “law firm.”
20. The computer-readable non-transitory medium of claim 15 , wherein further information is presented on the computer screen configured as a text script to be read by the agent.