Somshubra Majumdar
Somshubra Majumdar
Can you print out the system details - pytorch, Nemo, Numba versions. Seems to be the Numba kernels call for spec augment, but there shouldn't be a reason for it...
Hmm, ivr never tried it with Numba 0.56. I will try it locally later this month, I'm on vacation till the 19th.
Interesting. Also, numba just had a release of 0.56.2 which said it had cuda function caching bugfixes. Doubtful it's your case but could you try it ?
Ok I'll take a look once I'm back from vacation
This should now be fixed on main and the next release (r1.12.0)
We don't track val_loss for RNNT. Switch to metric of val_wer instead
Unless the script supports every possible combinations of models arch x CTC/RNNT x Char/subword, it should go in the subdirectory of examples/asr rather than at the root of examples/asr
It's supposed to be a script rathee than example, I see now. I'd advise to use a subdir inside scripts/speech_recognition anyway since it works to find confidence per word
No.move it to a subdir inside scripts/asr
No.move it to a subdir inside scripts/speech_recognition/confidence