Yujia SUN

wenetspeech occupy two much disk space even working on subset S

22

Hi, I am working on icefall wenetspeech egs. When processing the wenetspeech dataset, I found that all of the raw audio feature will be saved to disk which may take...

About the performance of the chunk-hopping mechanism

1

Have you done some evaluation of the chunk-hopping encoder. How about the performance?

Question about chunk mask

1

In the chunk based streaming stretagy, encoder mask is caculated by method "adaptive_enc_mask". I try to reproduce the mask which is showed as the fig. As the figure shows, the...

Dear friends, just asking if you have ever found any implementation of MWER in tensorflow?

1

I am working on MWER to improve transformer-based ASR model. That would be helpful, if you can share any experience. Thank you so much.