kafan1986

Facing same issue with latest version 5.0.27

> Seems trivial enough. The trivial solution is to just use masking (elemwise operations). This ensures that it's entirely differentiable. However if you have a link to the calculation of...

@chewxy Any update regarding this operation? I am stuck with one of my project as I am unable to integrate onnx model using cgo, so depending on this project.

Any update?

> Sorry... my internet at home was giving me a lot of grief. Turns out our national broadband network cannot really support a majority of people working from home at...

> Did you train this on your own data, and do you have Conformer-based baselines that don't show these issues, or at least that the issues are significantly less frequent...

> impact I have used Efficient conformer for testing conformer ASR model[https://github.com/burchim/EfficientConformer](url) and it gives better accuracy than vanilla conformer implementation. The loss used was RNNT for efficient conformer model....

> > > If you could take examples that have problematic deletions and shift the input by 1 or 2 or 4 frames, and see whether the deletions still appear,...

> Any update on this issue? I am planning to work on this issue and shall share any updates accordingly.

@SonakshiChauhan Can you share the link to download indic_timit_dataset? I can not find it online.