| New file |
| | |
|---|
| | | 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/asr_speaker_demo.wav' |
|---|
| | | output_dir = "./results" |
|---|
| | | inference_pipeline = pipeline( |
|---|
| | | task=Tasks.auto_speech_recognition, |
|---|
| | | model='damo/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn', |
|---|
| | | model_revision='v0.0.2', |
|---|
| | | vad_model='damo/speech_fsmn_vad_zh-cn-16k-common-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) |
|---|
| | | |
|---|
