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

update streaming paraformer text process

Browse Source
This commit is contained in:
haoneng.lhn 2023-05-04 16:40:34 +08:00
parent 5fc6f7899a
commit 297fafd674
1 changed files with 7 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
funasr/bin/asr_inference_paraformer_streaming.py
View File

@ -239,7 +239,7 @@ class Speech2Text:
feats_len = torch.tensor([feats_chunk2.shape[1]])
results_chunk2 = self.infer(feats_chunk2, feats_len, cache)
return ["".join(results_chunk1 + results_chunk2)]
return [" ".join(results_chunk1 + results_chunk2)]
results = self.infer(feats, feats_len, cache)
@ -299,12 +299,13 @@ class Speech2Text:
# Change integer-ids to tokens
token = self.converter.ids2tokens(token_int)
token = " ".join(token)
if self.tokenizer is not None:
text = self.tokenizer.tokens2text(token)
else:
text = None
results.append(text)
#if self.tokenizer is not None:
# text = self.tokenizer.tokens2text(token)
#else:
# text = None
results.append(token)
# assert check_return_type(results)
return results