-
```
import nemo.collections.asr as nemo_asr
asr_model = nemo_asr.models.EncDecCTCModel.from_pretrained("path_to_model.nemo")
transcript = asr_model.transcribe(paths=["path_to_audio.wav"], batch…
-
**My observation**
- With https://github.com/ys-zong/VLGuard/blob/main/VLGuard_eval.py, I am able to reproduce results not too far from Table 2 for **VLGuard dataset**.
- However, **I cannot reprodu…
-
MRE
```
program test
implicit none
integer :: a(:)
end program
```
GFortan error
```
(lf) lordansh@LordAnsh:~/dev/lfort1/lfortran$ gfortran try.f90
try.f90:3:15:
3 | integer :: a(…
-
Can you please specify the exact Pretrained Wav2Vec2 weights used and where to place them to make the script work? Thanks
-
This is much required feature for anyone to understand how openmp pass works and therefore I'll ask @adit4443ya to prioritise it before getting on to implement `collapse` in openmp pass. This will hel…
-
Towards #4397
MRE :
```fortran
module inf_mod_1
implicit none
private
public::is_finite
interface is_finite
module procedure is_finite_1
end interface is_finite
contains
subroutine …
-
Hi
can this be adopted for ASR (trained with CE loss) task ?
-
## Minimal Reproducible Example
```f90
integer function sub1()
implicit none
integer :: g
integer :: sub2
g = 2
sub1 = sub2(g)
end function
integer function sub2(x)
…
-
### Model description
Please add support for [moonshine](https://huggingface.co/UsefulSensors/moonshine) ASR models. The recent [github commit](https://github.com/usefulsensors/moonshine/commit/888f7…
-
Thank you for providing the code, but I encountered a problem while running it. I hope you can help me solve it. I used the 'lockdown' method to run the 'fmnist' dataset according to your code, but th…