problem on generate only noise

Question

problem on generate only noise

wl3b10s opened this issue 4 years ago · comments

with default setting,

simply run train on VCTK without extra global condition by :
python train.py

then
python generate.py --wav_out_path=testresult3900.wav --samples 320000 logdir/train/2020-10-22T13-35-57/model.ckpt-3900

seems doesn't generate voice wav but pure noise.

is there some more setting or condition needs to be added?

thanks.

Tanzy · Answer 1 · Sat Oct 24 2020 14:59:25 GMT+0800 (China Standard Time)

I'm also facing the same issue. I kept the model training for somewhere less than a week.

:) · Answer 2 · Wed Jul 28 2021 15:10:48 GMT+0800 (China Standard Time)

I took like 30~ hours to train 9000 steps and this is what came out: https://vocaroo.com/1kYnhSXQBGaN

This haunts me in my dreams. Jokes aside, the generated audio linked in the README was generated with 80k Steps, so I wouldn't expect 3900 steps to suffice

Marcos Serafim · Answer 3 · Sun Sep 19 2021 06:15:48 GMT+0800 (China Standard Time)

@PascalSchimmler thank you for the information. I was looking for some notion of how many steps is necessary to create good outputs.

magiccpp · Answer 4 · Wed May 10 2023 12:52:22 GMT+0800 (China Standard Time)

This is my trained model with 80000 steps,
https://drive.google.com/file/d/1MjCVFNw8XS0HgmPD0fthscxLMdGtB2cr/view?usp=share_link
The generated wave file:
generated_seed.wav.gz

ZhangXiaoying0116 · Answer 5 · Sun May 28 2023 03:08:44 GMT+0800 (China Standard Time)

@smartestrobotdai
Can you share your trained checkpoint with me?
I can't find this link https://drive.google.com/file/d/1MjCVFNw8XS0HgmPD0fthscxLMdGtB2cr/view?usp=share_link