speech_recognition/FunASR
1
0
Fork 0
mirror of https://github.com/modelscope/FunASR synced 2025-09-15 14:48:36 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

test

Browse Source
This commit is contained in:
语帆 2024-02-28 17:10:49 +08:00
parent e59ec16e6a
commit e0fca115cb
1 changed files with 2 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
funasr/models/lcbnet/model.py
View File

@ -443,8 +443,8 @@ class LCBNet(nn.Module):
encoder_out = encoder_out[0]
ocr_list_new = [[x + 1 if x != 0 else x for x in sublist] for sublist in ocr_sample_list]
ocr = torch.tensor(ocr_list_new)
ocr_lengths = ocr.new_full([1], dtype=torch.long, fill_value=ocr.size(1))
ocr = torch.tensor(ocr_list_new).to(device=kwargs["device"])
ocr_lengths = ocr.new_full([1], dtype=torch.long, fill_value=ocr.size(1)).to(device=kwargs["device"])
ocr, ocr_lens, _ = self.text_encoder(ocr, ocr_lengths)
fusion_out, _, _, _ = self.fusion_encoder(encoder_out,None, ocr, None)
encoder_out = encoder_out + fusion_out