Shinji Watanabe

Thanks for your report. @simpleoier, can you answer it for me?

We have some experience on using espnet models in the other framework, e.g., PyTorch Lightning. We're preparing some examples (but not very soon).

Many thanks for your report. @wyh2000, can you help Shuichiro?

> Thanks, my objective is not just to use it but to execute all the steps myself to understand and improve upon it. Also, I think it's important to make...

Thanks for posting the question. I did not notice the use of TensorRT in espnet. It would be great if someone who has such experience share the experience.

Thanks a lot, @Masao-Someki! @pyf98, can you review this PR?

Hmm, we seem to have some mysterious doc error. https://github.com/espnet/espnet/actions/runs/9191766239/job/25278895509?pr=5752 @cromz22, do you think this is specific to this PR or a general error?

Thanks for raising the issue. @Emrys365, can you answer it for me?

@kan-bayashi, can you help the issue?

> **Stage 5** In [these lines](https://github.com/espnet/espnet/blob/dbd73dd6d5fd8b672c3f7ff2bdd675ad8bd96477/egs2/TEMPLATE/asr2/asr2.sh#L843C1-L853C7), there are `${_suf}${dset}/${train_set}` and `${_suf}${dset}/${train_set}_sp`. Since `dset` comes from a previous for loop, I expect it should be removed. I think this sounds good...