egs_modelscope/asr/paraformer/speech_paraformer-tiny-commandword_asr_nat-zh-cn-16k-vocab544-pytorch/infer.py
@@ -20,7 +20,8 @@ task=Tasks.auto_speech_recognition, model="damo/speech_paraformer-tiny-commandword_asr_nat-zh-cn-16k-vocab544-pytorch", output_dir=output_dir_job, batch_size=64 batch_size=64, update_model=False, ) audio_in = os.path.join(split_dir, "wav.{}.scp".format(idx)) inference_pipeline(audio_in=audio_in)
