A system obtains conversation data corresponding to conversations between users and agents of a client. The system identifies a set of intents from the conversations and identifies a set of contexts, explicit elements, and implied elements of these intents. The system identifies actions that can be performed to recognize new explicit and implied elements from new conversations and to address intents in these new conversations. Based on these actions, the system generates a set of recommendations that can be provided to the client. As agents communicate with users, the system monitors adherence to the set of recommendations.

A system, computer program product, and method are disclosed. In an approach to train customer service agent using chatbots. The method includes training a chatbot for a customer chat simulation based on a customer service conversation data, a task scenario, and a customer persona. The method also includes monitoring an interaction between a customer service agent and the chatbot. The method further includes determining an assessment of the performance of the customer service agent based on the interaction between the customer service agent and the chatbot. The method additionally includes generating feedback for the customer service agent based on the assessment of the performance of the customer service agent.

Disclosed herein is a method for automated agent behavior recommendations for call quality improvement. The method performed at a server includes receiving a first data record and a second data record of a plurality of data records, each comprising communication between a first party and a second party and determining a first communication originated from the first party. The method includes determining a category for each section of the plurality of sections of the first communication, and a plurality of behavior distances between different categories associated with the plurality of sections. The method includes augmenting first metadata of the first data record and second metadata of the second data record to include associated behavior distances and determining an average performance ranking of the first party to generate a recommendation to increase the average performance ranking of the first party above a preconfigured threshold.

A system integrating a web browser and telephone is provided. A user enters, and the web browser receives, an input specifying a request to receive an audio output from the telephone. The web browser generates an audio stream which is communicated to the WebRTC gateway. The WebRTC gateway converts the audio stream from a first format into a second format. The WebRTC gateway communicates the converted audio stream to the SIP gateway, which forwards the audio stream to the telephone switch along with information identifying the telephone number to which the audio stream is directed. The telephone switch communicates the audio stream to the telephone where the audio stream is converted to be played using the speaker of the telephone.

A call center system for reducing communication latency includes an input/output (I/O) interface for receiving one or more queries from a customer terminal; a processor in communication with the I/O interface; and non-transitory computer readable media in communication with the processor that stores instruction code. The instruction code is executed by the processor and causes the processor to route the one or more queries to a plurality of artificial intelligent (AI) logic modules and receive, from one or more of the AI logic modules, information that facilitates providing, by a call center agent, responses to the one or more queries. The processor also routes actual responses to the one or more queries made by the call center agent to the AI logic modules; and receives from at least one AI logic module one or more scores associated with one or more metrics that rate different aspects if the actual responses. When at least one of the scores is below a threshold, the processor communicates training information to the call center agent. The training information is directed to a subject area associated with the score being below the threshold. Communication of the information to the call center agent reduces a latency between receipt of the one or more queries and communication of correct responses to the one or more queries.

A ranking of customer service interactions sessions that may benefit from supervisor input is provided. The customer service interactions sessions involve a Customer Service Representative (CSR), engaged in a customer service interaction with a customer. First, customer service interaction sessions, at a contact center server, between the CSRs and the customers begin. Data streams from CSR computer(s) to customer computer(s) are sent. Other data streams from the customer computers are received. The data streams are analyzed by a supervisor recommendation engine. Based on the analysis, the supervisor recommendation engine generates a ranking of customer service interactions sessions that would benefit most from supervisor input.

Methods for routing customers to an agent include receiving a customer communication; representing the customer communication as an array of one or more agent skills desired to handle the customer communication in one hot coding format or as a vector with an induced metric using an embedding algorithm; routing the represented customer communication to an agent having the one or more agent skills; measuring performance of the agent in relation to the one or more agent skills during or after the customer communication; updating, in real-time, one or more performance scores of the agent in a skill profile, wherein the one or more performance scores are related to the one or more agent skills; and routing subsequent customer communications based on the updated one or more performance scores.

A heuristic cross-selling recommendation engine includes capabilities to collect an unstructured data set and a current business context to suggest marketing actions. By providing a heuristic algorithm and executing the algorithm within the engine with the data set allows determination of predicted future contexts and optimal marketing actions. Such heuristic algorithms may learn from past marketing transactions and appropriate correlations with events and available data.

A system and method configured to generate a simulated caller dialog including a caller intended issue for a scenario for testing a customer service representative (CSR). A simulated caller dialog is presented to the CSR and a CSR response to the simulated caller dialog is received and includes a CSR interpretation of the caller intended issue to the simulated caller dialog. An understanding determination result based on an intent determination recognition score is generated by an intent determination recognition model is generated in response to a comparison of the CSR interpretation of the caller intended issue matching the caller intended issue in the simulated caller dialog. A CSR score is generated for the scenario based on the understanding determination result. The CSR score is recorded to a database.

A computer-implemented method for improving performance of a service representative that provides services. The method comprises determining a performance indicator representing performance of the service representative and if the performance indicator meets a condition, starting a computing process on a computing device to interact with the service representative in order to improve the performance of the service representative.