C.X. Gan

Thanks for your hard work on audio embedding. when I extract feature on my own dataset, I am wondering if the pretrained model can only accept the audio of which...

Dataset

1

Thanks for your hard work. Have you found that there are several audios that lasts more than 4sec? And another issue is that there are some format inconsistencies that may...

ThinResNet34

2

Thanks for your open-source code. I found that your ThinResNet34 implementation is different from that in your paper. I wonder if you have a revised version.

https://github.com/Annusha/temperature_schedules/blob/3ac1db63ffeb000d917c596be51d6120c19aa633/train_simCLR.py#L399 Thanks for your work. It looks like the temperature is a scalar and only related to the number of epoch. I wonder if my understanding is correct.

Thanks for your interesting work. I plan to train the model on Voxceleb2 but I encounter the warning like: UserWarning: leftover:50 in sampler, epoch:0, gpu:0, cropping.. warnings.warn( I wonder if...