MuseBox

Audio Analysis

The AI & ML models for Audio Analysis are responsible for the segmentation, classification and recognition of audio sources that are present in our processing data. 

It is also possible to generate synthetic music from real audio.

The information extracted could be directly logged, used or tagged as metadata.

Possible applications are (in combination with the other components of MuseBox):

  • Scene Detection and Classification
  • Gaming
  • Advertisement Recognition
  • Speech and Music segmentation
  • NLP
  • Behavioural Analysis

Supported Tasks

Audio Clustering
It recognizes an audio in a stream from a given database

Audio Filtering
It filters audio removing noise or background / foreground signals 

Audio Scene Classification
It classifies what a potential set could be for a given audio

Audio Segmentation
It segments different audios that compose a stream