Pre Trained Model

Question

Pre Trained Model

brappazzo opened this issue 7 years ago · comments

Does anyone have a pre trained model they could share? Specifically one trained on music (piano) data?

DiyuanLu · Answer 1 · Fri Dec 08 2017 23:53:36 GMT+0800 (China Standard Time)

This is a model trained with 72000+ steps with original configuration using VCTK corpus. Still far from fully trained, but could save sometime.
only changed SILENCE_TRESHOLD = 0.1 other than original 0.3
2017-12-04T13-48-11.zip

copy the model to the path "logdir/train/" and run it by type
python3 train.py --data_dir=corpus --gc_channels=32 --restore_from logdir/train/2017-12-04T13-48-11

Skylar Wurster · Answer 2 · Wed Mar 07 2018 08:20:30 GMT+0800 (China Standard Time)

@DiyuanLu what was the loss you were getting for this model? Just curious - I'm around 40k steps and its between 1.7 and 2.5. I think I need to adjust some parameters.

Zhaoyu Zhang · Answer 3 · Wed Mar 07 2018 21:40:34 GMT+0800 (China Standard Time)

@DiyuanLu How is your result of the final trained model? Does SILENCE_TRESHOLD = 0.1 work better than the original 0.3?

DiyuanLu · Answer 4 · Tue Mar 13 2018 17:50:55 GMT+0800 (China Standard Time)

Hi, for me it worked better. The threshold is basically a control of how much audio data you want to use to train your model. The higher, maybe the data is less but the quality maybe better. The lower, the more training data but a lot of them maybe non-speech noise. You have to find a trade-off, I guess.

…

@DiyuanLu How is your result of the final trained model? Does SILENCE_TRESHOLD = 0.1 work better than the original 0.3? -- You are receiving this because you were mentioned. Reply to this email directly or view it on GitHub: #307 (comment)

DiyuanLu · Answer 5 · Tue Mar 13 2018 17:54:01 GMT+0800 (China Standard Time)

@skywolf829 , I think the loss I got is similar. later around 60K, I think I got around 1.2.