This project combines a few really awesome open source projects to bring state-of-the-art speech recognition to the web. It uses the collects and qualifies the audio on the client-side using Voice Activity Detection This is my new favourite package! and then sends it to the Python/Flask server for processing with OpenAI Whisper for processing.
I wanted this project to remain two main files, one for front-end web code WITHOUT a framework and one for back-end Python code WITHOUT bloat. I think this is a good way to keep the code simple and easy to understand.
PRs are welcome!
-
Clone the repo
-
Install the dependencies
pip install -r requirements.txt -
Run the server
python server.py