Method and apparatus for providing active call guidance to an agent in a call center environment
A method and apparatus for guiding an agent during an active call between the customer and an agent is provided. The method comprises extracting, at a call analytics server (CAS), from a transcribed text of an audio of a call between a customer and an agent, a call context. Based on at least one of the call context, a call metadata, or a customer historical data, occurrence of a qualifying event is determined. Upon occurrence of the qualifying event, an action from multiple actions is identified based on at least one of the call context, the call metadata, or the customer historical data. The identified action is sent as guidance from the CAS to a graphical user interface (GUI) accessible by the agent, while the call is active.
1. A method for guiding an agent during an active call between the customer and an agent, the method comprising:
extracting, at a call analytics server (CAS), from a transcribed text of an audio of a call between a customer and an agent, a call context;
determining, at the CAS, based on at least one of the call context, a call metadata, or a customer historical data, whether a qualifying event has occurred, wherein the call metadata is received from a call metadata source remote to the CAS, and wherein the customer historical data is extracted from a feedback module;
upon occurrence of the qualifying event, identifying, by the CAS, an action from a plurality of actions, wherein the identifying is based on at least one of the call context, the call metadata, or the customer historical data, and wherein the plurality of actions are stored in an action repository remote to the CAS, wherein the identifying is performed using a trained artificial intelligence/machine learning (AI/ML) model, wherein the AI/ML model is trained to generate an output indicating at least one action based on an input of at least one of the call context, the call metadata, or the customer historical data; and
sending, from the CAS to a graphical user interface (GUI) accessible by the agent, the action as guidance, while the call is active.
2. The method of claim 1 , wherein the call context comprises at least one of a call intent, a customer sentiment, or a call entity.
3. The method of claim 2 , wherein the extracting is performed using a trained artificial intelligence/machine learning (AI/ML) model using natural language processing (NLP) techniques.
4. The method of claim 1 , further comprising:
receiving, at the CAS from the GUI, feedback regarding the action;
if the feedback is positive, preferring the action at a next instance of the identifying step; and
if the feedback is negative, deferring the action at a next instance of the identifying step.
5. An apparatus for guiding an agent during an active call between the customer and an agent, the apparatus comprising:
at least one processor;
a memory communicably coupled to the at least one processor, the memory comprising computer executable instructions, which when executed by the at least one processor perform a method comprising:
extracting, at a call analytics server (CAS), from a transcribed text of an audio of a call between a customer and an agent, a call context,
determining, at the CAS, based on at least one of the call context, a call metadata, or a customer historical data, whether a qualifying event has occurred, wherein the call metadata is received from a call metadata source remote to the CAS, and wherein the customer historical data is extracted from a feedback module,
upon occurrence of the qualifying event, identifying, by the CAS, an action from a plurality of actions, wherein the identifying is based on at least one of the call context, the call metadata, or the customer historical data, and wherein the plurality of actions are stored in an action repository remote to the CAS, wherein the identifying is performed using a trained artificial intelligence/machine learning (AI/ML) model, wherein the AI/ML model is trained to generate an output indicating at least one action based on an input of at least one of the call context, the call metadata, or the customer historical data, and
sending, from the CAS to a graphical user interface (GUI) accessible by the agent, the action as guidance, while the call is active.
6. The apparatus of claim 5 , wherein the call context comprises at least one of a call intent, a customer sentiment, or a call entity.
7. The apparatus of claim 6 , wherein the extracting is performed using a trained artificial intelligence/machine learning (AI/ML) model using natural language processing (NLP) techniques.
8. The apparatus of claim 5 , wherein the method further comprises:
receiving, at the CAS from the GUI, feedback regarding the action;
if the feedback is positive, preferring the action at a next instance of the identifying step; and
if the feedback is negative, deferring the action at a next instance of the identifying step.
9. A non-transitory computer-readable medium comprising computer-executable instructions, which when executed by at least one processor perform a method for guiding an agent during an active call between the customer and an agent, the method comprising:
extracting, at a call analytics server (CAS), from a transcribed text of an audio of a call between a customer and an agent, a call context;
determining, at the CAS, based on at least one of the call context, a call metadata, or a customer historical data, whether a qualifying event has occurred, wherein the call metadata is received from a call metadata source remote to the CAS, and wherein the customer historical data is extracted from a feedback module;
upon occurrence of the qualifying event, identifying, by the CAS, an action from a plurality of actions, wherein the identifying is based on at least one of the call context, the call metadata, or the customer historical data, and wherein the plurality of actions are stored in an action repository remote to the CAS, wherein the identifying is performed using a trained artificial intelligence/machine learning (AI/ML) model, wherein the AI/ML model is trained to generate an output indicating at least one action based on an input of at least one of the call context, the call metadata, or the customer historical data; and
sending, from the CAS to a graphical user interface (GUI) accessible by the agent, the action as guidance, while the call is active.
10. The computer-readable medium of claim 9 , wherein the call context comprises at least one of a call intent, a customer sentiment, or a call entity.
11. The computer-readable medium of claim 10 , wherein the extracting is performed using a trained artificial intelligence/machine learning (AI/ML) model using natural language processing (NLP) techniques.
12. The computer-readable medium of claim 9 , further comprising:
receiving, at the CAS from the GUI, feedback regarding the action;
if the feedback is positive, preferring the action at a next instance of the identifying step; and
if the feedback is negative, deferring the action at a next instance of the identifying step.