Speech2Text

This application has been produced during the internship of @kaicheng96. It aims at providing a proof of concept of features extraction from audio records.

Description

The application is splitted in two parts: a RShiny web interface and a Python core engine.

Features:

• Select a language between french, english, chinese;
• Choose an engine between Google API and a speechbrain model;
• Choose a file in wav OR record with your microphone;
• Run Speech2Text model;
• Run a model to add punctuation;
• Run a model to analyze emotions from audio record.

Visuals

Installation

Python requirements

TODO

R requirements

TODO

Install application

TODO

Run app Shiny

TODO

Authors and acknowledgment

Thank you @kaicheng96 for all the good work you have done during your 6 months intership.