best-rq-pytorch icon indicating copy to clipboard operation
best-rq-pytorch copied to clipboard

Published 20 hours ago • verified publisher icon lucasnewman

Missing Convolution Subsampling?

Open fmac2000 opened this issue 1 year ago • 1 comments

Hi Lucas, I'm looking over the code and I believe you have missed the two convolution subsampling layers in conformer.py,

4.1.1. NON-STREAMING MODELS The model has two convolution layers at the bottom which provide 4 times temporal-dimension reduction for the input sequences. The rest of the layers are a stack of Conformer models. We explore 0.6B model size which is extensively studied in the previous works. The model contains 24 layers of Conformer models.

Screenshot 2023-09-28 at 18 47 25

If you'd like I can create a pull request and implement this for you now. Thanks - If I've misunderstood the paper, please call me out! 😅

fmac2000 avatar Sep 28 '23 17:09 fmac2000