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 paraformer-speaker pipeline

Browse Source
This commit is contained in:
shixian.shi 2023-10-10 15:06:09 +08:00
parent 8a0930d682
commit ac6afabdd1
2 changed files with 20 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
egs_modelscope/asr_vad_spk/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/demo.py Normal file
View File

@ -0,0 +1,17 @@
from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks
if __name__ == '__main__':
audio_in = 'https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/vad_example.wav'
output_dir = "./results"
inference_pipeline = pipeline(
task=Tasks.auto_speech_recognition,
model='damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn',
vad_model='damo/speech_fsmn_vad_zh-cn-16k-common-pytorch',
#punc_model='damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch',
punc_model='damo/punc_ct-transformer_cn-en-common-vocab471067-large',
output_dir=output_dir,
)
rec_result = inference_pipeline(audio_in=audio_in, batch_size_token=5000, batch_size_token_threshold_s=40, max_single_segment_time=6000)
print(rec_result)

7
funasr/bin/asr_inference_launch.py
View File

@ -787,7 +787,7 @@ def inference_paraformer_vad_speaker(
time_stamp_writer: bool = True, time_stamp_writer: bool = True,
punc_infer_config: Optional[str] = None, punc_infer_config: Optional[str] = None,
punc_model_file: Optional[str] = None, punc_model_file: Optional[str] = None,
sv_model_file: Optional[str] = None, sv_model_file: Optional[str] = "~/.cache/modelscope/hub/damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/campplus_cn_common.bin",
streaming: bool = False, streaming: bool = False,
embedding_node: str = "resnet1_dense", embedding_node: str = "resnet1_dense",
sv_threshold: float = 0.9465, sv_threshold: float = 0.9465,
@ -933,7 +933,7 @@ def inference_paraformer_vad_speaker(
##### speaker_verification ##### ##### speaker_verification #####
################################## ##################################
# load sv model # load sv model
sv_model_dict = torch.load(sv_model_file, map_location=torch.device('cpu')) sv_model_dict = torch.load(sv_model_file.replace("~", os.environ['HOME']), map_location=torch.device('cpu'))
sv_model = CAMPPlus() sv_model = CAMPPlus()
sv_model.load_state_dict(sv_model_dict) sv_model.load_state_dict(sv_model_dict)
sv_model.eval() sv_model.eval()
@ -1084,7 +1084,6 @@ def inference_paraformer_vad_speaker(
logging.info("decoding, utt: {}, predictions: {}".format(key, text_postprocessed_punc)) logging.info("decoding, utt: {}, predictions: {}".format(key, text_postprocessed_punc))
torch.cuda.empty_cache() torch.cuda.empty_cache()
distribute_spk(asr_result_list[0]['sentences'], sv_output) distribute_spk(asr_result_list[0]['sentences'], sv_output)
import pdb; pdb.set_trace()
return asr_result_list return asr_result_list
return _forward return _forward
@ -2030,7 +2029,7 @@ def inference_launch(**kwargs):
return inference_paraformer(**kwargs) return inference_paraformer(**kwargs)
elif mode == "paraformer_streaming": elif mode == "paraformer_streaming":
return inference_paraformer_online(**kwargs) return inference_paraformer_online(**kwargs)
elif mode == "paraformer_vad_speaker": elif mode.startswith("paraformer_vad_speaker"):
return inference_paraformer_vad_speaker(**kwargs) return inference_paraformer_vad_speaker(**kwargs)
elif mode.startswith("paraformer_vad"): elif mode.startswith("paraformer_vad"):
return inference_paraformer_vad_punc(**kwargs) return inference_paraformer_vad_punc(**kwargs)