Agent coaching using voice services
Systems and methods for providing intelligent assistance using voice services for agent coaching. In some embodiments, an Information Handling System (IHS) may include: a processor; and a memory coupled to the processor, the memory having program instructions stored thereon that, upon execution, cause the IHS to: receive an instruction from a voice service provider in response to audio captured during a communication between a customer and an agent, where the instruction includes a sentiment indication; identify a resource used by the agent during the communication; determine a result of the communication; and store the sentiment indication, the resource identification, and the result of the communication, in a coaching database.
1. An Information Handling System (IHS), comprising:
a processor;
a microphone; and
a memory coupled to the processor, the memory having program instructions stored thereon that, upon execution, cause the IHS to:
receive an instruction from a voice service provider in response to audio captured by the microphone during a communication between a customer and an agent, wherein the instruction includes sentiment indications;
automatically display the sentiment indications to the support agent in a graphical user interface, wherein the sentiment indications show a level for each sentiment during the communication;
display a list of resources that are relevant to the communication in the graphical user interface, wherein the resources are identified using a ranking algorithm using machine learning techniques;
identify a resource used by the agent during the communication;
determine a result of the communication; and
store the sentiment indication, a resource identification associated with the identified resource, and the result of the communication, in a coaching database.
2. The IHS of claim 1 , wherein the program instructions, upon execution, further cause the IHS to obtain context information of a product during the communication.
3. The IHS of claim 2 , wherein the instruction is generated, at least in part, by brokering the use of different language models, wherein at least one of the different language models is selected based upon the context information.
4. The IHS of claim 1 , wherein the instruction includes a text portion corresponding to the audio.
5. The IHS of claim 4 , wherein the instruction includes a keyword spoken during the communication.
6. The IHS of claim 1 , wherein the sentiment indication corresponds to: the customer or the agent.
7. The IHS of claim 1 , wherein the sentiment indication is selected from the group consisting of: anger, joy, and hesitation.
8. The IHS of claim 7 , wherein the sentiment indication includes a confidence score.
9. The IHS of claim 1 , wherein the resource includes at least one of: a registration record, a warranty record, a support history, or a knowledgebase article.
10. The IHS of claim 1 , wherein the result of the communication is selected from the group consisting of: positive outcome, positive experience, negative outcome, and negative experience.
11. The IHS of claim 1 , wherein the program instructions, upon execution, further cause the IHS to:
coach another agent during another communication with another customer using the coaching database.
12. The IHS of claim 11 , wherein to coach the other agent, the program instructions, upon execution, further cause the IHS to:
identify the resource, to the other agent, for making available to the other customer during the other communication, at least in part, in response to a comparison between: (i) another sentiment indication received from the voice service provider in response to other audio captured during the other communication, and (ii) the stored sentiment indication.
13. The IHS of claim 1 , wherein the program instructions, upon execution, further cause the IHS to:
route another communication to another agent selected, at least in part, based upon a comparison between another instruction received from the voice service provider in connection with the other communication, and the instruction.
14. A hardware memory device having program instructions stored thereon that, upon execution by a processor of an Information Handling System (IHS), cause the IHS to:
receive an instruction from a voice service provider in response to audio captured by a microphone coupled to the IHS during an ongoing communication between a customer and an agent, wherein the instruction includes: (i) text recognized by brokering the use of different language models selected using context information obtained during the ongoing communication, and (ii) more than one sentiment indications;
display the more than one sentiment indications to the support agent in a graphical user interface, wherein the sentiment indications show a level for each sentiment during the communication; and
identify a resource, to the agent, for making available to the customer during the communication, at least in part, in response to a comparison between the sentiment indications and a stored sentiment indication associated with the resource, wherein the resource is identified using a ranking algorithm using machine learning techniques.
15. The hardware memory device of claim 14 , wherein the resource is identified based upon a service tag uttered during the ongoing communication.
16. The hardware memory device of claim 14 , wherein the resource is identified based upon a brand, a service, or a product name uttered during the ongoing communication.
17. The hardware memory device of claim 14 , wherein the resource is identified among a plurality of other resources ranked by historical sentiment change.
18. A method, comprising:
receiving an instruction from a voice service provider in response to audio captured by a microphone coupled to an IHS during a prior communication between a customer and an agent, wherein the instruction includes sentiment indications;
identifying a resource used by the agent during the prior communication;
determining a result of the prior communication;
storing the sentiment indications, a resource identification associated with the identified resource, and the result of the prior communication, in a coaching database;
receiving another instruction from the voice service provider in response to audio captured during an ongoing communication between another customer and another agent, wherein the other instruction includes current sentiment indications;
automatically displaying the current sentiment indications to the support agent in a graphical user interface, wherein the sentiment indications show a level for each sentiment during the communication; and
identifying the resource, to the other agent, for making available to the other customer during the ongoing communication, at least in part, in response to a comparison between the current sentiment indications and the sentiment indications stored in the coaching database, wherein the resource is identified using a ranking algorithm using machine learning techniques.
19. The method of claim 18 , wherein the resource is identified based upon a brand, a service, or a product name uttered during the ongoing communication and recognized by brokering the use of different language models selected using context information.
20. The method of claim 18 , wherein the resource is identified among a plurality of other resources ranked by predicted sentiment change.