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

decoding

Browse Source
This commit is contained in:
游雁 2024-06-13 17:40:40 +08:00
parent c553a8db17
commit 7355e20503
1 changed files with 7 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
funasr/models/llm_asr/model.py
View File

@ -496,8 +496,11 @@ class LLMASR2(nn.Module):
batch_size, frames, _ = speech.shape batch_size, frames, _ = speech.shape
with torch.cuda.amp.autocast(enabled=False):
# audio encoder # audio encoder
encoder_out, encoder_out_lens = self.audio_encoder(speech.permute(0, 2, 1), speech_lengths) encoder_out, encoder_out_lens = self.audio_encoder(
speech.permute(0, 2, 1), speech_lengths
)
# audio_adaptor # audio_adaptor
encoder_out, encoder_out_lens = self.audio_adaptor(encoder_out, encoder_out_lens) encoder_out, encoder_out_lens = self.audio_adaptor(encoder_out, encoder_out_lens)