Abstract: A method and an apparatus for determining speaker behavior in a conversation in a call comprising an audio is provided. The apparatus includes a call analytics server comprising a processor and a memory, which performs the method. The method comprises receiving, at a call analytics server (CAS), a call audio comprising a speech of a first speaker, identifying an emotion based on the speech and identifying a sentiment based on a call text corresponding to the speech. Based on the identified emotion and sentiment, a behavior of the first speaker in the conversation is determined.

Abstract: A method and an apparatus for predicting satisfaction of a customer pursuant to a call between the customer and an agent is provided. The method comprises receiving a transcribed text of the call, dividing the transcribed text into a plurality of phases of a conversation, extracting at least one call feature for each of the plurality of phases, receiving call metadata, extracting metadata features from the call metadata, combining the call features and the metadata features, and generating an output, using a trained machine learning (ML) model, based on the combined features, indicating whether the customer is satisfied or not. The ML model is trained to generate an output indicating whether the customer is satisfied or not, based on an input of the combined features.

Abstract: A method and an apparatus for predicting satisfaction of a customer pursuant to a call between the customer and an agent, in which the method comprises receiving a transcribed text of the call, dividing the transcribed text into a plurality of phases of a conversation, extracting at least one call feature for each of the plurality of phases, receiving call metadata, extracting metadata features from the call metadata, combining the call features and the metadata features, and generating an output, using a trained machine learning (ML) model, based on the combined features, indicating whether the customer is satisfied or not. The ML model is trained to generate an output indicating whether the customer is satisfied or not, based on an input of the combined features.

Abstract: Systems and methods for identifying sensitive content in electronic files are disclosed. In an embodiment, a request is received to determine whether an electronic file contains sensitive content. The electronic file is preprocessed based on a file type of the electronic file, resulting in a first input file and a second input file. The first input file is inputted to a first machine learning engine, which classifies the first input file for numerical items in the sensitive content. The second input file is inputted to a second machine learning engine, which classifies the second input file for textual items the sensitive content. A report is generated based on a combination of a first output from the first machine learning engine and a second output from the second machine learning engine, where the report indicates items of the sensitive content that are contained in the electronic file.

Abstract: Systems and methods for identifying sensitive content in electronic files are disclosed. In an embodiment, a request is received to determine whether an electronic file contains sensitive content. The electronic file is preprocessed based on a file type of the electronic file, resulting in a first input file and a second input file. The first input file is inputted to a first machine learning engine, which classifies the first input file for numerical items in the sensitive content. The second input file is inputted to a second machine learning engine, which classifies the second input file for textual items the sensitive content. A report is generated based on a combination of a first output from the first machine learning engine and a second output from the second machine learning engine, where the report indicates items of the sensitive content that are contained in the electronic file.