update paraformer-speaker pipeline

shixian.shi

2023-10-10 ac6afabdd1e8843a12b7c66dd73538eea461a79a

update paraformer-speaker pipeline

 egs_modelscope/asr_vad_spk/speech_paraformer-large-vad-punc-spk_asr_nat-zh-cn/demo.py

New 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)


 funasr/bin/asr_inference_launch.py

@@ -787,7 +787,7 @@
        time_stamp_writer: bool = True,
        punc_infer_config: 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,
        embedding_node: str = "resnet1_dense",
        sv_threshold: float = 0.9465,
@@ -933,7 +933,7 @@
            #####  speaker_verification  #####
            ##################################
            # 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.load_state_dict(sv_model_dict)
            sv_model.eval()
@@ -1084,7 +1084,6 @@
            logging.info("decoding, utt: {}, predictions: {}".format(key, text_postprocessed_punc))
        torch.cuda.empty_cache()
        distribute_spk(asr_result_list[0]['sentences'], sv_output)
        import pdb; pdb.set_trace()
        return asr_result_list

    return _forward
@@ -2030,7 +2029,7 @@
        return inference_paraformer(**kwargs)
    elif mode == "paraformer_streaming":
        return inference_paraformer_online(**kwargs)
    elif mode == "paraformer_vad_speaker":
    elif mode.startswith("paraformer_vad_speaker"):
        return inference_paraformer_vad_speaker(**kwargs)
    elif mode.startswith("paraformer_vad"):
        return inference_paraformer_vad_punc(**kwargs)