sf

Thank you , very much. I have obtained it. And by the way , I find the last wav in the test will be ignored during decoding.

Thank you, I find it's work well. I want to use the model trained by librispeech as the source domain, and train an accent ASR model using small data. Now,...

In the git: https://github.com/tramphero/kaldi , the example of librispeech

I also find this problem

I find it in the git https://github.com/tramphero/kaldi .

> Please run test.py outside the pytorch_binding dir. It still failed.

Thank you very much!