hifigan v1 gradient explosion ？

[unparalleled-ysj]

The number of training steps can be seen from the figure

Some data will also appear a straight line after 100k in the v2 version. Do you have any suggestions?

[ZDisket]

There was a similar issue a bit ago and I replied with this: #387 (comment). Have you tried it?

[unparalleled-ysj]

There was a similar issue a bit ago and I replied with this: #387 (comment). Have you tried it?

Okay, let me test it. By the way, should the parameter discriminator_train_start_steps be set to 100k or set to 0 like you

[ZDisket]

@unparalleled-ysj Try 0 first.

[jarred1989]

@ZDisket @unparalleled-ysj Any good news? I ran into the same problem, so I changed the optimizer and set discriminator_train_start_steps to 0 according to #387, the synthesized waveforms are no longer silence, but they are even worse than mb-melgan.

[ZDisket]

@jarred1989 A while back I came into the conclusion that there is just something wrong with TensorflowTTS HiFi-GAN that can't be alleviated with simple optimizer or config changes, so I settled with adding the MPD to MB-MelGAN, which trains slower but helps increase multispeaker performance, and may make it possible to successfully finetune another model to adapt it to a target speaker, as I am testing out.

[jarred1989]

@ZDisket Thanks for your opinion, I may try it out later.

[chenht2021]

make learning rate smaller will be good. I tried on two corpus male and female, use default config, failed; smaller the lr, the result is good.

[ZDisket]

@chenht2010 What learning rate did you use?

[chenht2021]

only changed the generator:
generator_optimizer_params:
values: [0.000125, 0.000125, 0.000625, 0.000625, 0.0000625, 0.00003125, 0.000015625, 0.000001]

[dathudeptrai]

only changed the generator:
generator_optimizer_params:
values: [0.000125, 0.000125, 0.000625, 0.000625, 0.0000625, 0.00003125, 0.000015625, 0.000001]

can you make a pull request to change the default config for better training :D ?

[chenht2021]

only changed the generator:
generator_optimizer_params:
values: [0.000125, 0.000125, 0.000625, 0.000625, 0.0000625, 0.00003125, 0.000015625, 0.000001]

can you make a pull request to change the default config for better training :D ?

sure

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs.