From a05e753d11d9c36983ec4e58c421dbcf86d1dcd4 Mon Sep 17 00:00:00 2001 From: Xian Shi <40013335+R1ckShi@users.noreply.github.com> Date: 星期二, 17 十月 2023 16:47:27 +0800 Subject: [PATCH] Merge branch 'main' into dev_onnx --- egs_modelscope/asr_vad_spk/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/demo.py | 17 +++++++++++++++++ 1 files changed, 17 insertions(+), 0 deletions(-) diff --git a/egs_modelscope/asr_vad_spk/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/demo.py b/egs_modelscope/asr_vad_spk/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/demo.py new file mode 100644 index 0000000..7b2bd18 --- /dev/null +++ b/egs_modelscope/asr_vad_spk/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/demo.py @@ -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/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) + -- Gitblit v1.9.1