Subject: [AUDITORY] Announcing Asteroid: PyTorch-based audio source separation toolkit for researchers From: Manuel Pariente <pariente.mnl@xxxxxxxx> Date: Wed, 7 Oct 2020 20:00:43 +0200--00000000000030e75a05b1187f96 Content-Type: text/plain; charset="UTF-8" === Apologies for cross-posting === Dear list, We are happy to present Asteroid, the PyTorch-based audio source separation toolkit for researchers. Asteroid provides code to train state-of-the-art deep learning models on common audio source separation and speech enhancement datasets. Asteroid's most noticeable features: - Asteroid comes with architectures such as Deep Clustering, Chimera++, TasNet, ConvTasNet, DualPathRNN, DualPathTransformer, DCUNet, DCCRN. - Permutation invariant training is made extremely simple with our PITLossWrapper. - Asteroid supports several filterbanks, losses, metrics and datasets for easy experimentations. - Our model hub hosted on Zenodo enables us to share and download pretrained models. - All the code is released under the MIT License. *Webpage *: https://asteroid-team.github.io/ *Code : *https://github.com/mpariente/asteroid *Docs* : https://mpariente.github.io/asteroid/ *Model hub *: https://zenodo.org/communities/asteroid-models *Paper* : https://arxiv.org/abs/2005.04132 *Presentation video *: https://www.youtube.com/watch?v=imnZxQwuNcg We hope that this resource will be useful to the community, and we invite any one of you to help us grow it in the directions you would like to see. All the best, Manuel Pariente --00000000000030e75a05b1187f96 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable <div dir=3D"ltr"><p dir=3D"ltr" style=3D"line-height:1.2;text-align:justify= ;margin-top:0pt;margin-bottom:0pt"><span style=3D"font-family:arial,sans-se= rif"><span style=3D"font-size:10pt;color:rgb(0,0,0);background-color:transp= arent;font-weight:400;font-style:normal;font-variant:normal;text-decoration= :none;vertical-align:baseline;white-space:pre-wrap">=3D=3D=3D Apologies for= cross-posting =3D=3D=3D</span></span></p><p dir=3D"ltr" style=3D"line-heig= ht:1.656;text-align:justify;margin-top:12pt;margin-bottom:12pt"><span style= =3D"font-family:arial,sans-serif"><span style=3D"font-size:10pt;color:rgb(0= ,0,0);background-color:transparent;font-weight:400;font-style:normal;font-v= ariant:normal;text-decoration:none;vertical-align:baseline;white-space:pre-= wrap">Dear list,</span></span></p><p style=3D"line-height:1.656;text-align:= justify;margin-top:12pt;margin-bottom:12pt"><span style=3D"font-family:aria= l,sans-serif"><span style=3D"font-size:10pt;color:rgb(0,0,0);background-col= or:transparent;font-weight:400;font-style:normal;font-variant:normal;text-d= ecoration:none;vertical-align:baseline;white-space:pre-wrap">We are happy t= o present Asteroid, the PyTorch-based audio source separation toolkit for r= esearchers. Asteroid provides code to train state-of-the-art deep learning = models on common audio source separation and speech enhancement datasets.<b= r></span></span></p><p style=3D"line-height:1.656;text-align:justify;margin= -top:12pt;margin-bottom:12pt"><span style=3D"font-family:arial,sans-serif">= <span style=3D"font-size:10pt;color:rgb(0,0,0);background-color:transparent= ;font-weight:400;font-style:normal;font-variant:normal;text-decoration:none= ;vertical-align:baseline;white-space:pre-wrap">Asteroid's most noticeab= le features: <br></span></span></p><ul><li><span style=3D"font-family:arial= ,sans-serif"><span style=3D"font-size:10pt;color:rgb(0,0,0);background-colo= r:transparent;font-weight:400;font-style:normal;font-variant:normal;text-de= coration:none;vertical-align:baseline;white-space:pre-wrap">Asteroid comes = with architectures such as Deep Clustering, Chimera++, TasNet, ConvTasNet, = DualPathRNN, DualPathTransformer, DCUNet, DCCRN.</span></span></li><li><spa= n style=3D"font-family:arial,sans-serif"><span style=3D"font-size:10pt;colo= r:rgb(0,0,0);background-color:transparent;font-weight:400;font-style:normal= ;font-variant:normal;text-decoration:none;vertical-align:baseline;white-spa= ce:pre-wrap">Permutation invariant training is made extremely simple with o= ur PITLossWrapper.<br></span></span></li><li><span style=3D"font-family:ari= al,sans-serif"><span style=3D"font-size:10pt;color:rgb(0,0,0);background-co= lor:transparent;font-weight:400;font-style:normal;font-variant:normal;text-= decoration:none;vertical-align:baseline;white-space:pre-wrap">Asteroid supp= orts several filterbanks, losses, metrics and datasets for easy experimenta= tions.</span></span></li><li><span style=3D"font-family:arial,sans-serif">O= ur model hub hosted on Zenodo enables us to share and download pretrained m= odels.</span></li><li><span style=3D"font-family:arial,sans-serif">All the = code is released under the MIT License. <br></span></li></ul><p style=3D"li= ne-height:1.656;text-align:justify;margin-top:12pt;margin-bottom:12pt"><spa= n style=3D"font-family:arial,sans-serif"><span style=3D"font-size:10pt;colo= r:rgb(0,0,0);background-color:transparent;font-weight:400;font-style:normal= ;font-variant:normal;text-decoration:none;vertical-align:baseline;white-spa= ce:pre-wrap"><b>Webpage </b>: <a href=3D"https://asteroid-team.github.io/" = target=3D"_blank">https://asteroid-team.github.io/</a><br><b>Code : </b><a = href=3D"https://github.com/mpariente/asteroid" target=3D"_blank">https://gi= thub.com/mpariente/asteroid</a><b><br>Docs</b> : <a href=3D"https://mparien= te.github.io/asteroid/" target=3D"_blank">https://mpariente.github.io/aster= oid/</a><br><b>Model hub </b>: <a href=3D"https://zenodo.org/communities/as= teroid-models" target=3D"_blank">https://zenodo.org/communities/asteroid-mo= dels</a><br><b>Paper</b> : <a href=3D"https://arxiv.org/abs/2005.04132" tar= get=3D"_blank">https://arxiv.org/abs/2005.04132</a><br><b>Presentation vide= o </b>: <a href=3D"https://www.youtube.com/watch?v=3DimnZxQwuNcg" target=3D= "_blank">https://www.youtube.com/watch?v=3DimnZxQwuNcg</a></span></span></p= ><p style=3D"line-height:1.656;text-align:justify;margin-top:12pt;margin-bo= ttom:12pt"><span style=3D"font-family:arial,sans-serif"><span style=3D"font= -size:10pt;color:rgb(0,0,0);background-color:transparent;font-weight:400;fo= nt-style:normal;font-variant:normal;text-decoration:none;vertical-align:bas= eline;white-space:pre-wrap">We hope that this resource will be useful to th= e community, and we invite any one of you to help us grow it in the directi= ons you would like to see.<br></span></span></p><p style=3D"line-height:1.6= 56;text-align:justify;margin-top:12pt;margin-bottom:12pt"><span style=3D"fo= nt-family:arial,sans-serif"><span style=3D"font-size:10pt;color:rgb(0,0,0);= background-color:transparent;font-weight:400;font-style:normal;font-variant= :normal;text-decoration:none;vertical-align:baseline;white-space:pre-wrap">= All the best, <br>Manuel Pariente<br></span></span></p></div> --00000000000030e75a05b1187f96--