The repository adapts the example from https://github.com/vivjay30/clearbuds. It uses the same models and the example use case from 'cascaded.sh'.
conda create -n clearbud_env python=3.8
conda activate clearbud_env
pip3 install torch torchvision torchaudio
cd clearbuds
pip install -r requirements.txt
- Download the clearbud repo: https://drive.google.com/file/d/1C1m7UjZRNVq84jzx49krJ5tsaaMwalRG/view?usp=sharing
- Copy the models: final.pth.tar and model.pt to a folder called 'models' in this clearbuds directory.
python3 main.py --dir_data "/path/to/data_directory" --left_channel 0 , --right_channel 1