ml-audio-enhancemants.md

ML audio enhancements

Overview

Create an audio enhancing feature for Jitsi Meet using available ML models.

Description

Noise / audio problems plague video conferences. This project aims to integrate existing ML models in an "audio enhancing" feature which would make voice be clearer and made front and center, clean of background noise.

Expected outcomes

A new feature for the web application, using existing ML models, improving audio quality in usual circumstances like barking dogs, police cars, etc.

Skills

JavaScript, React, TensorFlow.

Possible mentors

saghul, hristoterezov.

Expected project size

350 hours.

Difficulty

Hard.