Virtual coaching assistant

A non-transitory computer-readable medium comprising instructions that, when executed, cause operations comprising:

• training a plurality of machine learning models based at least in part on a plurality of first feature vectors comprising a plurality of features and a first plurality of corresponding values, the plurality of features comprising a first feature related to a scenario classification, a second feature related to subject demographic information, a third feature to emotional characteristics, and a fourth feature related to an interaction outcome;
• receiving a request to generate a coaching suggestion from a user device associated with a first user;
• generating a data set associated with the first user and a second user using data comprising interactions between the first user and the second user stored in a datastore, wherein the data set comprises a second feature vector including the plurality of features comprising a first feature related to a scenario classification, a second feature related to subject demographic information, a third feature to emotional characteristics, and a fourth feature related to an interaction outcome and a second plurality of corresponding values;
• feeding the data set to a first machine learning model of the plurality of machine learning models, wherein the first machine learning model is configured to predict a scenario associated with the first user and the second user, wherein the scenario comprises at least one of an agent scenario, a customer scenario, or agent coaching;
• determining, based on the scenarios, to import a second machine learning model of the plurality of machine learning models;
• feeding the data set and scenario to the second machine learning model, wherein the second machine learning model is configured to determine second model output;
• feeding the second model output to a third machine learning model of the plurality of machine learning models, wherein the third machine learning model is configured to determine third model output;
• generating, based on combining the second model output from the second machine learning model and the third model output from the third machine learning model, a coaching suggestion associated with the first user communicating with the second user, wherein the second model output comprises suggested content of the coaching suggestion, and wherein the third model output comprises suggested diction of the coaching suggestion;
• transmitting, to the user device, the coaching suggestion;
• and
• outputting the coaching suggestion via a display or an audio component of the user device.

A virtual coaching assistant comprising:

• one or more processors;
• and
• memory storing instructions that, when executed by the one or more processors, cause the virtual coaching assistant to:
• train a plurality of machine learning models based at least in part on a plurality of first feature vectors comprising a plurality of features and a first plurality of corresponding values, the plurality of features comprising a first feature related to a scenario classification, a second feature related to subject demographic information, a third feature to emotional characteristics, and a fourth feature related to an interaction outcome;
• receive a request to generate a coaching suggestion from a user device associated with a first user;
• generate a data set associated with the first user and a second user using data comprising interactions between the first user and the second user stored in a datastore, wherein the data set comprises a second feature vector including the plurality of features comprising a first feature related to a scenario classification, a second feature related to subject demographic information, a third feature to emotional characteristics, and a fourth feature related to an interaction outcome and a second plurality of corresponding values;
• feed the data set to a first machine learning model wherein the first machine learning model is configured to predict a scenario associated with the first user and the second user, wherein the scenario comprises at least one of an agent scenario, a customer scenario, or agent coaching;
• determine, based on the scenario, to import a second machine learning model of the plurality of machine learning models;
• feed the data set and scenario to the second machine learning model, wherein the second machine learning model is configured to determine second model output;
• feed the second model output to a third machine learning model of the plurality of machine learning models, wherein the third machine learning model is configured to determine third model output;
• generate, based on combining the second model output from the second machine learning model and the third model output from the third machine learning model, a coaching suggestion associated with the first user communicating with the second user, wherein the second model output comprises suggested content of the coaching suggestion, and wherein the third model output comprises suggested diction of the coaching suggestion;
• transmit, to the user device, the coaching suggestion;
• and
• output the coaching suggestion via a display or an audio component of the user device.