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

refine decoding process, merge flow and vocoder

Browse Source
This commit is contained in:
志浩 2024-07-10 12:47:25 +08:00
parent 8335712c66
commit d64ba60f40
1 changed files with 1 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
funasr/models/llm_asr/model.py
View File

@ -2383,6 +2383,7 @@ class LLMASR5(nn.Module):
seq_input[0, prompt_len + out_token_len, :] = self.codec_embedder(top_ids)[0]
out_token_len += 1
out_tokens = out_tokens[:, :out_token_len, :]
if decoding_length is None:
return out_tokens
else: