This tool automatically gets lyrics of songs from audio files using the OpenAI Whisper model or API.
-
Switch to either the
model
orapi
folder:cd model
-
Copy the config file:
cp config_dist.py config.py
Make sure to modify the config values as appropriate.
-
Create a virtual environment using venv for example:
python -m venv .env
-
Install requirements:
source .env/bin/activate pip install -r requirements.txt
Run:
python transcribe.py
- If you get the error:
See: https://stackoverflow.com/questions/52805115/certificate-verify-failed-unable-to-get-local-issuer-certificate
urllib.error.URLError: <urlopen error [SSL: CERTIFICATE_VERIFY_FAILED] certificate verify failed: self signed certificate in certificate chain (_ssl.c:1002)
- If you get the error:
See: openai/whisper#1027
RuntimeError: Model has been downloaded but the SHA256 checksum does not not match. Please retry loading the model.
Please fork this repo and make a pull request to either the transcribe
or test
branch.