fmac2000

Int4 Support

6

Hello Authors, I apologise for asking questions unrelated to an issue with the repo however, would you consider support a newer paradigm I came across whilst reading a recent [paper](https://www.researchgate.net/publication/367557918_Understanding_INT4_Quantization_for_Transformer_Models_Latency_Speedup_Composability_and_Failure_Cases)?...

Missing Convolution Subsampling?

1

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...

Pretrained weights + Inference Code

3

Hello Authors, Fantastic work, will you be releasing the source code to the wider community? Thank you for all your hard work - well done!