Skip to content

Latest commit

 

History

History
13 lines (11 loc) · 863 Bytes

README.md

File metadata and controls

13 lines (11 loc) · 863 Bytes

VerbaLex

Automatic Speech Recognition system for non-native English speakers. This project is a university final-year project. Inspired by Google's Project Euphonia, this project aims to replicate a similar machine-learning model for more people.

Setup

Environment

The environment for this project uses Anaconda. To set up the anaconda environment, follow the guide on PyCharm's documentation about configuring a conda environment.

Development

This project is currently developed by using OpenAI's Whisper model from HuggingFace. The project will fine-tune the model based on different accents from different countries.