System and method of diarization and labeling of audio data

A system for diarization and labeling of audio data, the system comprising:

• an audio database server comprising a plurality of audio files;
• a speech-to-text (SST) server that:
• receives a set of audio data from the audio database server, wherein each audio data is an audio recording of a conversation between two or more speakers, wherein at least one of the two or more speakers is a customer service agent, further wherein at least one of the two or more speakers is a customer,
• analyzes the audio data to create textual transcripts for each the audio data, wherein the textual transcripts include an identification of homogeneous speaker segments;
• a processor that:
• receives a sub-set of the textual transcripts and a sub-set of audio data for diarization from the SST server, wherein the sub-set of audio data is the audio data associated with each of the sub-set of the textual transcripts,
• performs a blind diarization on the sub-set of textual transcripts to segment and cluster the sub-set of textual transcripts into a plurality of textual speaker clusters, wherein the number of textual speaker clusters is at least equal to a number of speakers in each textual transcript,
• applies at least one heuristic to the textual speaker clusters with a processor to select textual speaker clusters likely to be associated with an identified group of speakers,
• analyzes the selected textual speaker clusters to extract a subset of the textual speaker clusters known to be spoken by the identified group of speakers,
• creates at least one linguistic model based on the extracted subset of textual speaker clusters;
• and
• an audio source that provides new transcribed audio data to the processor;
• wherein the processor applies the at least one linguistic model to a new audio file transcript from an audio source to perform diarization of the new audio file by blind diarizing the new audio file, comparing each new textual speaker cluster to the at least one linguistic model, and labeling each textual speaker cluster as belonging to a customer service agent or belonging to a customer.