Skip to content

Latest commit

 

History

History
39 lines (39 loc) · 4.38 KB

PRETRAINED.md

File metadata and controls

39 lines (39 loc) · 4.38 KB
Raw

Pretrained Models

Model Dataset Example
DeepClustering WSJ0-2mix model = DeepClustering.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
DeepClustering WSJ0-3mix model = DeepClustering.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)
DANet WSJ0-2mix model = DANet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
DANet WSJ0-3mix model = DANet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)
DANet (fixed attractor) WSJ0-2mix model = FixedAttractorDANet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
DANet (fixed attractor) WSJ0-3mix model = FixedAttractorDANet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)
DANet LibriSpeech model = DANet.build_from_pretrained(task="librispeech", sample_rate=16000, n_sources=2)
ADANet WSJ0-2mix model = ADANet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
ADANet WSJ0-3mix model = ADANet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)
LSTM-TasNet WSJ0-2mix model = LSTMTasNet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
LSTM-TasNet WSJ0-3mix model = LSTMTasNet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)
Conv-TasNet WSJ0-2mix model = ConvTasNet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
Conv-TasNet WSJ0-3mix model = ConvTasNet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)
Conv-TasNet MUSDB18 model = ConvTasNet.build_from_pretrained(task="musdb18", sample_rate=44100)
Conv-TasNet WHAM model = ConvTasNet.build_from_pretrained(task="wham/separate-noisy", sample_rate=8000)
Conv-TasNet WHAM model = ConvTasNet.build_from_pretrained(task="wham/enhance-single", sample_rate=8000)
Conv-TasNet WHAM model = ConvTasNet.build_from_pretrained(task="wham/enhance-both", sample_rate=8000)
Conv-TasNet LibriSpeech model = ConvTasNet.build_from_pretrained(task="librispeech", sample_rate=16000, n_sources=2)
DPRNN-TasNet WSJ0-2mix model = DPRNNTasNet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
DPRNN-TasNet WSJ0-3mix model = DPRNNTasNet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)
DPRNN-TasNet LibriSpeech model = DPRNNTasNet.build_from_pretrained(task="librispeech", sample_rate=16000, n_sources=2)
MMDenseLSTM MUSDB18 model = MMDenseLSTM.build_from_pretrained(task="musdb18", sample_rate=44100, target="vocals")
MMDenseLSTM (bass, drums, other, vocals) MUSDB18 model = ParallelMMDenseLSTM.build_from_pretrained(task="musdb18", sample_rate=44100)
Open-Unmix MUSDB18 model = OpenUnmix.build_from_pretrained(task="musdb18", sample_rate=44100, target="vocals")
Open-Unmix (bass, drums, other, vocals) MUSDB18 model = ParallelOpenUnmix.build_from_pretrained(task="musdb18", sample_rate=44100)
Open-Unmix MUSDB18-HQ model = OpenUnmix.build_from_pretrained(task="musdb18hq", sample_rate=44100, target="vocals")
Open-Unmix (bass, drums, other, vocals) MUSDB18-HQ model = ParallelOpenUnmix.build_from_pretrained(task="musdb18hq", sample_rate=44100)
DPTNet WSJ0-2mix model = DPTNet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
DPTNet WSJ0-3mix model = DPTNet.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)
CrossNet-Open-Unmix MUSDB18 model = CrossNetOpenUnmix.build_from_pretrained(task="musdb18", sample_rate=44100)
D3Net MUSDB18 model = D3Net.build_from_pretrained(task="musdb18", sample_rate=44100, target="vocals")
D3Net (bass, drums, other, vocals) MUSDB18 model = ParallelD3Net.build_from_pretrained(task="musdb18", sample_rate=44100)
D3Net MUSDB18-HQ model = D3Net.build_from_pretrained(task="musdb18hq", sample_rate=44100, target="vocals")
D3Net (bass, drums, other, vocals) MUSDB18-HQ model = ParallelD3Net.build_from_pretrained(task="musdb18hq", sample_rate=44100)
SepFormer WSJ0-2mix model = SepFormer.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=2)
SepFormer WSJ0-3mix model = SepFormer.build_from_pretrained(task="wsj0-mix", sample_rate=8000, n_sources=3)