QuartzNet-ASR-pytorch icon indicating copy to clipboard operation
QuartzNet-ASR-pytorch copied to clipboard

verified publisher icon Kirili4ik
→

Metadata

Automatic Speech Recognition (ASR) model QuartzNet trained on English CommonVoice. In PyTroch with CTC loss and beam search.

Results 1 QuartzNet-ASR-pytorch issues
Sort by recently updated
recently updated
newest added

Why using separable conv in C1 and C2 instead of normal conv1d?

3 comment

Thank you for sharing your great work. I noticed you have used `sepconv_bn` in C1 and C2 instead of `conv_bn_act`. Is it on purpose? Does it give better results? https://github.com/Kirili4ik/QuartzNet-ASR-pytorch/blob/ec6073ef76d1ce0419bc62065ec746cb12a63efc/model.py#L49

mohsen-goodarzi avatar mohsen-goodarzi

← Metadata

15
Stars
2
Forks
Watchers

Owner

verified publisher icon Kirili4ik

Metadata

Automatic Speech Recognition (ASR) model QuartzNet trained on English CommonVoice. In PyTroch with CTC loss and beam search.

Back