Video call routing and management based on artificial intelligence determined facial emotion
A video stream of a video call between a communication endpoint of customer and a communication endpoint of a contact center agent is received. The video stream of the video call is processed in real-time to generate a real-time emotion transcript. The real-time emotion transcript tracks a plurality of separate emotions based on non-verbal expressions (e.g. facial expressions) that occur in the video stream of the video call. For example, different emotions of both the customer and the contact center agent may tracked in the real-time emotion transcript. The real-time emotion transcript is compared to an emotion transcript of at least one previous video call to determine if the video call should be handled differently in the contact center. In response to determining that video call should be handled differently in the contact center, an action determined to change how the video call is managed in the contact center.
1. A first contact center comprising:
a microprocessor; and
a computer readable medium, coupled with the microprocessor and comprising microprocessor readable and executable instructions that, when executed by the microprocessor, cause the microprocessor to:
receive a video stream of a real-time video call between a communication endpoint of a customer and a first communication endpoint of a first contact center agent;
process, in real-time, the video stream of the real-time video call to generate a first real-time emotion transcript, wherein the first real-time emotion transcript tracks a plurality of separate emotions based on a plurality of non-verbal expressions that occur in the video stream of the real-time video call;
compare the first real-time emotion transcript to a second emotion transcript of at least one previous video call to determine if the real-time video call should be handled differently in the first contact center, wherein the comparison comprises determining at least one pattern in the first real-time emotion transcript and determining that one or more of the at least one pattern is similar to a pattern in the second emotion transcript; and
in response to determining that the real-time video call should be handled differently in the first contact center, determine an action to change, during the real-time video call, how the real-time video call is managed in the first contact center.
2. The first contact center of claim 1 , wherein the plurality of non-verbal expressions comprise at least one micro-expression that cannot be visually detected by a human without the aid of the microprocessor.
3. The first contact center of claim 1 , wherein the plurality of separate emotions are emotions of the customer and emotions of the first contact center agent.
4. The first contact center of claim 2 , wherein a duration of one or more of the at least one micro-expression is about 1/30 of a second.
5. The first contact center of claim 1 , wherein the first real-time emotion transcript comprises a first real-time customer emotion transcript and a first real-time agent emotion transcript, wherein the second emotion transcript comprises a plurality of prior customer emotion transcripts and a plurality of prior agent emotion transcripts of a plurality of prior video calls, and wherein comparing the first real-time emotion transcript to the second emotion transcript comprises identifying at least one of the plurality of prior video calls that has a customer emotion transcript that is similar to the first real-time customer emotion transcript and an agent emotion transcript that is similar to the first real-time agent emotion transcript.
6. The first contact center of claim 1 , wherein the first real-time emotion transcript comprises a first real-time customer emotion transcript of a customer, wherein the second emotion transcript comprises a second customer emotion transcript of the customer.
7. The first contact center of claim 1 , wherein the first real-time emotion transcript and the second emotion transcript are transcripts associated with a specific type of real-time video call and wherein the specific type of real-time video call is associated with at least one of: a supported type of service provided by the first contact center, a contact center queue, a supported product, and a skill of an agent.
8. The first contact center of claim 1 , wherein the first real-time emotion transcript comprises a first real-time customer emotion transcript and a first real-time agent emotion transcript that are combined to form a composite transcript.
9. The first contact center of claim 1 , wherein the plurality of non-verbal expressions comprises at least one facial expression and at least one gesture that each occur in the video stream of the real-time video call.
10. The first contact center of claim 1 , wherein the first real-time emotion transcript is a composite emotion transcript of all participants in the real-time video call and wherein the second emotion transcript is a composite emotion transcript of all participants in the at least one previous video call.
11. A method comprising:
receiving, by a microprocessor in a first contact center, a video stream of a real-time video call between a communication endpoint of a customer and a first communication endpoint of a first contact center agent;
processing, in real-time, by the microprocessor, the video stream of the real-time video call to generate a first real-time emotion transcript, wherein the first real-time emotion transcript tracks a plurality of separate emotions based on a plurality of non-verbal expressions that occur in the video stream of the real-time video call;
comparing, by the microprocessor, the first real-time emotion transcript to a second emotion transcript of at least one previous video call to determine if the real-time video call should be handled differently in the first contact center, wherein the comparing comprises determining at least one pattern in the first real-time emotion transcript and determining that one or more of the at least one pattern is similar to a pattern in the second emotion transcript; and
in response to determining that the real-time video call should be handled differently in the first contact center, determining, by the microprocessor and during the real-time video call, an action to change how the real-time video call is managed in the first contact center.
12. The method of claim 11 , wherein the at least one pattern has at least a first emotion at a first level and at least a second emotion at a second level.
13. The method of claim 11 , wherein the first real-time emotion transcript comprises a first real-time customer emotion transcript related to a type of service and/or product, wherein the second emotion transcript comprises a plurality of prior customer emotion transcripts of a plurality of prior video calls, and wherein comparing the first real-time emotion transcript to the second emotion transcript comprises identifying at least one of the plurality of prior video calls that has a customer emotion transcript that is similar to the first real-time customer emotion transcript and that is related to the type of service and/or product.
14. The method of claim 12 , wherein the comparing comprises comparing the first level with a level of the first emotion in the second emotion transcript and comparing the second level with a level of the second emotion in the second emotion transcript.
15. The method of claim 11 , wherein the comparing comprises determining a variance of at least one emotion in the first real-time emotion transcript.
16. The method of claim 11 , wherein the comparing comprises determining a composite variance between at least two emotions in the first real-time emotion transcript.
17. A system comprising:
a microprocessor; and
a computer readable medium, coupled with the microprocessor and comprising microprocessor readable and executable instructions that, when executed by the microprocessor, cause the microprocessor to:
receive a video stream of a real-time video call between a first communication endpoint and a second communication endpoint;
process, in real-time, the video stream of the real-time video call to generate a first real-time emotion transcript, wherein the first real-time emotion transcript tracks a plurality of separate emotions based on a plurality of non-verbal expressions that occur in the video stream of the real-time video call;
comparing the first real-time emotion transcript to a second emotion transcript of at least one previous video call to determine, during the real-time video call, if the real-time video call should be handled differently, wherein the comparing comprises determining at least one pattern in the first real-time emotion transcript and determining that one or more of the at least one pattern is similar to a pattern in the second emotion transcript; and
in response to determining that the real-time video call should be handled differently, performing an action on the real-time video call.
18. The system of claim 17 , wherein the first real-time emotion transcript is generated based on a machine learning algorithm trained to identify emotions based on at least two micro-expressions.