CONVTASNET_BASE_LIBRI2MIX

torchaudio.pipelines.CONVTASNET_BASE_LIBRI2MIX

Pre-trained Source Separation pipeline with ConvTasNet [Luo and Mesgarani, 2019] trained on Libri2Mix dataset [Cosentino et al., 2020].

The source separation model is constructed by conv_tasnet_base() and is trained using the training script lightning_train.py here with default arguments.

Please refer to SourceSeparationBundle for usage instructions.