US-12620320-B1 - Contact center quality management control using generative artificial intelligence

US12620320B1US 12620320 B1US12620320 B1US 12620320B1US-12620320-B1

Abstract

A quality management output relative to agent performance during a contact center interaction is generated using a trained generative artificial intelligence model to automate the reporting and further action of such quality management output. Recording information associated with a contact center interaction between an agent and an end user is obtained. The recording information is processed using a generative artificial intelligence model to determine a quality management output for the contact center interaction. An agent report is then generated based on the quality management output and associated materials and provided to the agent, and, optionally, to a supervisor device associated with the agent. In some cases, a simulated contact center interaction may be generated to further train the agent based on the quality management output.

Inventors

Ryan Christopher Ang
Periyaven Naiken Gopalla

Assignees

ZOOM COMMUNICATIONS, INC.

Dates

Publication Date: 20260505
Application Date: 20240429

Claims (20)

1 . A method, comprising: generating, by a first large language model of a contact center system, a summary of a contact center interaction based on a transcription of the contact center interaction, wherein the contact center interaction is facilitated by the contact center system between an agent device and a user device and corresponds to a video communication modality or an asynchronous communication modality; determining, by a second large language model of the contact center system retrieving recording information associated with the contact center interaction from a data store in which the recording information is stored and processing the recording information, a quality management output for the contact center interaction, wherein the recording information includes the summary of the contact center interaction and indicates an evaluation of agent performance during the contact center interaction; generating, by the second large language model processing the quality management output, a simulated contact center interaction that includes simulated end user media to present to the agent device, wherein the simulated end user media includes video content configured for real-time response by an agent using the agent device and corresponds to the video communication modality; and outputting, to the agent device, an agent report that includes a hyperlink usable for the agent device to access the simulated contact center interaction and respond to the video content.
2 . The method of claim 1 , comprising: transmitting the agent report to a supervisor device associated with the agent.
3 . The method of claim 1 , wherein determining the quality management output for the contact center interaction comprises: extracting, using the second large language model, insights from the recording information; and determining the quality management output based on the insights.
4 . The method of claim 1 , wherein determining the quality management output for the contact center interaction comprises: evaluating, using the second large language model, agent performance during the contact center interaction using the recording information to determine a score for the agent; and determining the quality management output based on the score for the agent.
5 . The method of claim 1 , wherein outputting the agent report comprises: including a timeline of insights with one or more comments generated using the second large language model within the agent report.
6 . The method of claim 1 , wherein the contact center interaction is facilitated via a video conference implemented by a contact center as a service platform.
7 . The method of claim 1 , wherein the contact center interaction includes a chat message conversation or text message conversation and the recording information corresponds to messages of the chat message conversation or of the text message conversation.
8 . The method of claim 1 , comprising: filtering, by the second large language model, sensitive or private content of the recording information.
9 . A non-transitory computer readable medium storing instructions operable to cause one or more processors to perform operations comprising: generating, by a first large language model of a contact center system, a summary of a contact center interaction based on a transcription of the contact center interaction, wherein the contact center interaction is facilitated by the contact center system between an agent device and a user device and corresponds to a video communication modality or an asynchronous communication modality; determining, by a second large language model of the contact center system retrieving recording information associated with the contact center interaction from a data store in which the recording information is stored and processing the recording information, a quality management output for the contact center interaction, wherein the recording information includes the summary of the contact center interaction and indicates an evaluation of agent performance during the contact center interaction; generating, by the second large language model processing the quality management output, simulated contact center interaction that includes simulated end user media to present to the agent device, wherein the simulated end user media includes video content configured for real-time response by an agent using the agent device and corresponds to the video communication modality; and outputting, to the agent device, an agent report that includes a hyperlink usable for the agent device to access the simulated contact center interaction and respond to the video content.
10 . The non-transitory computer readable medium of claim 9 , wherein the quality management output is determined based on insights extracted from the recording information using the second large language model.
11 . The non-transitory computer readable medium of claim 9 , wherein the quality management output is determined based on a score determined by evaluating agent performance during the contact center interaction according to the recording information.
12 . The non-transitory computer readable medium of claim 9 , wherein the agent report is accessible to a supervisor device associated with the agent.
13 . The non-transitory computer readable medium of claim 9 , wherein the agent report visually represents comments generated using the second large language model in a timeline format.
14 . The non-transitory computer readable medium of claim 9 , wherein the second large language model filters sensitive or private content of the recording information.
15 . A system, comprising: a memory subsystem; and processing circuitry configured to execute instructions stored in the memory subsystem to: generate, by a first large language model of a contact center system, a summary of a contact center interaction based on a transcription of the contact center interaction, wherein the contact center interaction is facilitated by the contact center system between an agent device and a user device and corresponds to a video communication modality or an asynchronous communication modality; determine, by a second large language model of the contact center system retrieving recording information associated with the contact center interaction from a data store in which the recording information is stored and processing the recording information, a quality management output for the contact center interaction, wherein the recording information includes the summary of the contact center interaction and indicates an evaluation of agent performance during the contact center interaction; generate, by the second large language model processing the quality management output, a simulated contact center interaction that includes simulated end user media to present to the agent device, wherein the simulated end user media includes video content configured for real-time response by an agent using the agent device and corresponds to the video communication modality; and output, to the agent device, an agent report that includes a hyperlink usable for the agent device to access the simulated contact center interaction and respond to the video content.
16 . The system of claim 15 , wherein, to determine the quality management output for the contact center interaction, the processing circuitry is configured to execute the instructions to: determine the quality management output based on insights extracted using the second large language model from the recording information.
17 . The system of claim 15 , wherein, to determine the quality management output for the contact center interaction, the processing circuitry is configured to execute the instructions to: determine the quality management output based on a score determined for the agent by the second large language model evaluating agent performance during the contact center interaction using the recording information.
18 . The system of claim 15 , wherein the agent report includes a timeline of insights including one or more comments generated using the second large language model.
19 . The system of claim 15 , wherein the contact center interaction is facilitated using a synchronous communication service of a unified communications as a service software platform or of a contact center as a service platform.
20 . The system of claim 15 , wherein sensitive or private content of the recording information is filtered to determine the quality management output.

Description

CROSS-REFERENCE TO RELATED APPLICATION(S) This application claims the benefit of U.S. Provisional Application Ser. No. 63/530,837, filed Aug. 4, 2023, the entire disclosure of which is herein incorporated by reference. FIELD This disclosure generally relates to contact center quality management control, and, more specifically, to generating a quality management output relative to agent performance during a contact center interaction using a trained generative artificial intelligence (AI) model and automating the reporting and further action of such quality management output. BRIEF DESCRIPTION OF THE DRAWINGS This disclosure is best understood from the following detailed description when read in conjunction with the accompanying drawings. It is emphasized that, according to common practice, the various features of the drawings are not to-scale. On the contrary, the dimensions of the various features are arbitrarily expanded or reduced for clarity. FIG. 1 is a block diagram of an example of an electronic computing and communications system. FIG. 2 is a block diagram of an example internal configuration of a computing device of an electronic computing and communications system. FIG. 3 is a block diagram of an example of a software platform implemented by an electronic computing and communications system. FIG. 4 is a block diagram of an example of a contact center system. FIG. 5 is a block diagram of an example quality management control functionality of a contact center system. FIG. 6 is an illustration of a first example of contents included in an agent report generated based on a quality management output for a contact center interaction. FIG. 7 is an illustration of a second example of contents included in an agent report generated based on a quality management output for a contact center interaction. FIG. 8 is a flowchart of an example of a technique for contact center quality management control. DETAILED DESCRIPTION The use of contact centers by or for service providers is becoming increasingly common to address customer support requests over various modalities, including telephony, video, text messaging, chat, and social media. In one example, a contact center may be implemented by an operator of a software platform, such as a unified communications as a service (UCaaS) platform or a contact center as a service (CCaaS) platform, for a customer of the operator. Users of the customer may engage with the contact center to address support requests over one or more communication modalities enabled for use with the contact center by the software platform. In another example, the operator of such a software platform may implement a contact center to address customer support requests related to the software platform itself. A contact center may utilize one or more quality management mechanisms to validate that agents of the contact center are providing high quality service to its end users. Generally, quality management in the contact center context refers to undertakings for ensuring that customer interactions with a contact center meet a defined quality standard within all communication modalities and across all agents of the contact center. These undertakings result in high end user satisfaction with their contact center interactions, improved agent efficacy at addressing end user questions and requests, and previously unnoticed product and operational issues being identified. Conventional quality management approaches require user action, typically involving a supervisor or other managerial personnel manually defining a set of standards to which to hold their agents and thereafter evaluating agent performance by manually reviewing recordings of interactions between the agents and respective end users. While such approaches may yield satisfactory results, they suffer from drawbacks based on the required manual user action. For example, such manual approaches may often or at least from time to time fail to accurately consider or characterize certain agent activity or speech, whether in isolated circumstances or as a behavioral pattern. In doing so, conventional quality management approaches risk allowing false positives or false negatives that could otherwise be caught and addressed and which, by not being addressed, may result in decreased agent performance quality. In another example, such manual approaches require that the supervisor or other managerial personnel take the time necessary to produce written output for the subject agent to review and that they thereafter follow up with the agent to ensure that the agent has acted on that output. However, due to memory lapses, time constraints, or other factors, such written outputs may sometimes not be produced and thus agents may not receive quality management feedback at all. Implementations of this disclosure address problems such as these by generating a quality management output relative to agent performance during a contact center interaction and automat