egs_modelscope/asr/data2vec/speech_data2vec_pretrain-zh-cn-aishell2-16k-pytorch/infer_after_finetune.py
@@ -48,5 +48,5 @@ params["required_files"] = ["am.mvn", "decoding.yaml", "configuration.json"] params["output_dir"] = "./checkpoint" params["data_dir"] = "./data/test" params["decoding_model_name"] = "valid.cer_ctc.ave.pth" params["decoding_model_name"] = "valid.cer_ctc.ave.pb" modelscope_infer_after_finetune(params)
