# download the prerequisite asset models (lip regressor and wav2vec)
wget http://audio2photoreal_models.berkeleyvision.org/asset_models.tar
tar xvf asset_models.tar
rm asset_models.tar