FunASR/egs_modelscope/wenetspeech/paraformer

ModelScope: Paraformer-large Model

Highlight

ModelScope: Paraformer-Large Model

• Fast: Non-autoregressive (NAR) model, the Paraformer can achieve comparable performance to the state-of-the-art AR transformer, with more than 10x speedup.
• Accurate: SOTA in a lot of public ASR tasks, with a very significant relative improvement, capable of industrial implementation.
• Convenient: Quickly and easily download Paraformer-large from Modelscope for finetuning and inference.
  • Support finetuning and inference on AISHELL-1 and AISHELL-2.
  • Support inference on AISHELL-1, AISHELL-2, Wenetspeech, SpeechIO and other audio.

How to infer using a pretrained ModelScope Paraformer-large Model

Inference

• Setting parameters in paraformer_large_infer.sh
  • ori_data: please set the wenetspeech raw data path
  • data_dir: data output dictionary
  • exp_dir: the result path
  • model_name: speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch # base model, download from modelscope
  • test_sets: please set the testsets name

• Then you can run the pipeline to infer with:

  sh ./paraformer_large_infer.sh