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 2023-05-05 11:36:26 +08:00
parent 297fafd674
commit 7987a7b6fa
1 changed files with 2 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
funasr/bin/asr_inference_paraformer_streaming.py
View File

@ -301,10 +301,6 @@ class Speech2Text:
token = self.converter.ids2tokens(token_int) token = self.converter.ids2tokens(token_int)
token = " ".join(token) token = " ".join(token)
#if self.tokenizer is not None:
# text = self.tokenizer.tokens2text(token)
#else:
# text = None
results.append(token) results.append(token)
# assert check_return_type(results) # assert check_return_type(results)
@ -556,8 +552,8 @@ def inference_modelscope(
input_lens = torch.tensor([stride_size]) input_lens = torch.tensor([stride_size])
asr_result = speech2text(cache, raw_inputs[:, sample_offset: sample_offset + stride_size], input_lens) asr_result = speech2text(cache, raw_inputs[:, sample_offset: sample_offset + stride_size], input_lens)
if len(asr_result) != 0: if len(asr_result) != 0:
final_result += asr_result[0] final_result += " ".join(asr_result) + " "
item = {'key': "utt", 'value': [final_result]} item = {'key': "utt", 'value': [final_result.strip()]}
else: else:
input_lens = torch.tensor([raw_inputs.shape[1]]) input_lens = torch.tensor([raw_inputs.shape[1]])
cache["encoder"]["is_final"] = is_final cache["encoder"]["is_final"] = is_final
@ -751,12 +747,3 @@ def main(cmd=None):
if __name__ == "__main__": if __name__ == "__main__":
main() main()
# from modelscope.pipelines import pipeline
# from modelscope.utils.constant import Tasks
#
# inference_16k_pipline = pipeline(
# task=Tasks.auto_speech_recognition,
# model='damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch')
#
# rec_result = inference_16k_pipline(audio_in='https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav')
# print(rec_result)