From 9252c1d80a746d642f08dcdff512e872ed12fb96 Mon Sep 17 00:00:00 2001 From: dohe0342 Date: Tue, 11 Apr 2023 15:54:07 +0900 Subject: [PATCH] from local --- .../.data2vec_audio.py.swp | Bin 40960 -> 40960 bytes .../data2vec_audio.py | 2 +- 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/.data2vec_audio.py.swp b/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/.data2vec_audio.py.swp index e02802b80daaa152f84c3eb69812d6afd622c0e8..66d3265fab6da22401bd87291d5af70ae2b423b4 100644 GIT binary patch delta 116 zcmZoTz|?SnNi@kI%+puFQqO<^2m}}y*jY?dTsMlow`YCL%)p?%nbU!rQFR|11H*J6 zo(9C3K%4=@Ea;cd$XnEcjn0-qFE-NbQRg`@4AnXT}L4$GuPH|a-xUm JW+nG~5dg`!8}k4F delta 114 zcmZoTz|?SnNi@kI%+puFQqO<^2m}}y9x|Jz2yGO7Z_oOPnSsG*Gp7SLqv}yM28MY+ zJQs)yfH)tBy@8k=h!3zbFjNAu(Pm4>@61d}6_ea#H+#75W1O7mAv$@Lo8sgG5AMx= H?wcY2%hMa* diff --git a/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/data2vec_audio.py b/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/data2vec_audio.py index 6e73be508..ad9222398 100644 --- a/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/data2vec_audio.py +++ b/egs/librispeech/ASR/pruned_transducer_stateless_d2v_v2/data2vec_audio.py @@ -499,7 +499,7 @@ class Data2VecAudioModel(BaseFairseqModel): prompt_padding_mask = torch.zeros(prompt.size()[0], prompt.size()[1]).type(torch.BoolTensor).to(features.device) print(prompt_padding_mask.size()) print(padding_mask.size()) - padding_mask = torch.cat([prompt_padding_mask, padding_mask]) + padding_mask = torch.cat([prompt_padding_mask, padding_mask], dim=1) print(padding_mask.size()) features = self.layer_norm(features)