Voice classification
Clone the repo and install the project dependencies. It is recommended to use a virtual environment (e.g. conda)
git clone https://github.com/jerpint/voicemd
cd voicemd
pip install -e .
First, download the weights locally:
cd ~/voicemd/voicemd
curl -L -o model.pt "https://www.dropbox.com/s/r3ydpfxr3h4vlzg/best_model_commonvoice?dl=1"
Then run as a script:
from voicemd.predict import make_a_prediction
make_a_prediction(
sound_filepath='/path/to/your/sound.wav',
config_filepath='config.yaml',
best_model_path='model.pt'
)