P.S - Work in progress for major improvements

Audiolytics

Project repository

 

Audiolytics provides state of the art solutions to enhance audio analytics.

At the current stage, it's able to provide:

  • Gender classification: a custom machine learning model able to identify the speaker's gender from an audio file, from which audio features are extracted.
  • Transcript: leveraging state of the art Deep Learning APIs from Google to provide fast and accurate transcriptions.
  • Sentiment analysis: leveraging Huggingface's Transformer architecture to classify the dominant sentiment (positive or negative) in the transcript.