| CPC G09B 5/04 (2013.01) [G06Q 30/015 (2023.01)] | 18 Claims |
|
1. A computer-implemented method for customer service representative training via a conversation with a simulated customer using a computer system associated with a customer service representative, a chat engine, a text-to-speech engine, and a speech-to-text engine, wherein the computer system, the chat engine, the text-to-speech engine, and the speech-to-text engine are coupled to a communications network, the method comprising:
receiving, by the computer system, a profile of a simulated customer;
causing display, by the computer system, of the profile of the simulated customer;
generating, by the chat engine, a plurality of textual prompts associated with the simulated customer and based on the profile of the simulated customer;
generating, by the text-to-speech engine, a plurality of audio prompts associated with the simulated customer, each one of the plurality of audio prompts corresponding to one of the plurality of textual prompts, including adjusting one or more vocal characteristics of simulated speech in the plurality of audio prompts based upon information related to the voice of the simulated customer from the profile of the simulated customer;
playing, by the computer system, each one of the plurality of audio prompts;
receiving, by the computer system, a plurality of audio responses associated with the customer service representative, wherein each one of the plurality of audio responses is received in response to a corresponding one of the plurality of audio prompts;
generating, by the speech-to-text-engine, a plurality of textual responses associated with the customer service representative, wherein each one of the plurality of textual responses corresponds to one of the plurality of audio responses;
analyzing, by the computer system, the plurality of textual responses associated with the customer service representative to generate feedback for the customer service representative, including determining a sentiment associated with each of a plurality of words in each textual response and annotating each textual response using the determined sentiments; and
storing the plurality of textual prompts and the plurality of annotated textual responses in a database.
|