search

TaoRuijie / AVCleanse

ICASSP 2023: 'Speaker recognition with two-step multi-modal deep cleansing'
27 stars 3 forks source link

Question for cleasing step #2

Open CloudWalkerW opened 4 months ago

CloudWalkerW commented 4 months ago

Hello, I'm working on audio-visual speaker recognition system and I read your code.

I would like to ask if your cleasing code is included in this repo? Or that you only provided clean list for training data.

Cause I didn't find the part that you did cleasing steps. Does that means if your models in this repo are just simple speaker recognition models?

Thanks for your reply!

TaoRuijie commented 4 months ago

Hi,

Sorry, my original plan was to release everything while the cleansing code was too redundant. I attempted several ways but did not include it finally. So, as the introduction mentioned, "This project contains the code of audio-visual speaker recognition"

Thanks for your understanding.