Parcollet Titouan

HI, we are sorry for the delay. Lot of stuffs have changed since then and few conflicts exist. If you are ok and have time, could you make a new...

Hi, sorry for the late reply, we are very very busy with SpeechBrain and PyTorch-Kaldi won't be maintained anymore very soon. Unfortunately, no. And it won't.

Hi, sorry for the late reply, we are very very busy with SpeechBrain and PyTorch-Kaldi won't be maintained anymore very soon. Isn't the energy of the current frame also added...

This looks like an issue created by Pytorch 1.12 [see](https://github.com/pytorch/pytorch/issues/80809). As a verification, could you try to downgrade to PyTorch 1.11 and start again ?

Hi, I tried with a fresh install, and everything works fine. Hence I am not able to reproduce the error. It is most likely that this is due to a...

Hi, you can create a single module in the YAML as a Sequential, or Module List or whatever, and then give it to the checkpointer. See the asr_model variable here:...

Hi, are you training with the recommended yaml file ? Did you change anything ? How many GPUs are you using?

Ok, if you only use one GPU you will have to increase gradient_accumulation to compensate :-)

Alright, we never tried but it is feasible. You would have to 1. fine-tune a wav2vec 2.0 with BPE; 2. Train a transformer LM with the same BPE; 3. Combine...

We added many things to the next major version, so it will most likely be postponed to Fall. Note that a pyctcdecode integration would be extremely simple to do (literally...