Kelvin Qin

## Title - defect on CER ## Description I found two defects inside Supervised_trainer.py, which is all about CER --- (_train_epoches) line 267, cer = self.metric(targets, y_hats) the returned CER...

how to run decoding on chain model built by myself

2

Guenter, Thanks for your work, I successfully compiled and tested your code. And also I can do decoding using your pre-trained model (kaldi-generic-en-tdnn_f-r20190609) now. Great work! Meanwhile, I am learning...

Is it possible for me to import my BVH to play your humanoid avatar (EmilianaCecil)?

5

Dear Emiliana, Great work. I am working on speech-driven gesture generation (speech2bvh), if I want to try your tool to generate humanized animation using my bvh, not sure if it...

Overfitting? test AST on GTZAN

23

Hi Dear Gong Yuan, Thanks for your excellent work, I learnt a lot. After understanding your code (partially), I test it on GTZAN using my own training framework (Just copy...

quality of image is not as good as the one in the blog

3

Hi William, I am following your script to build cifar10 model, I checked the image after 20 epochs, which is not as good as the one, not sure if you...

Can you provide the training script to build the HuBERT-Discrete Content Encoder? Thanks!

1

Dear, thanks for your great work, can you point me how to rebuild HuBert-Discrete model?

Zhengxi, Thanks for your work, not sure if I can have the scripts and data file like phone.map under frontend directory? Many thanks! Kelvin

To repeat your result

5

Gheyret, Thanks for your work, I tried to reproduce your work, but after 300 epochs (batch size = 20), CER was still around 10%, not sure how many epochs for...

W2Llayer

2

Dear Gheyret, Thanks for your work. I spent some time today to try to figure out the source of this feature extraction layer, can you point me the paper/any reference...