Train ChatGPT with audio.
This component recognizes speech from your audio files and feeds it to a ChatGPT Retrieval Plugin instance.
This project has been generated with create_app.
- Place your mp3 files in audio_to_chatgpt/input
- In the variables.env file, put the Retrieval Plugin bearer token in the RETRIEVAL_PLUGIN_BEARER_TOKEN variable
- Make sure the Retrieval Plugin and its datastore are running
- Build the Docker image:
make docker_build
- Run the Docker image:
make docker_run