Nikita Kononov

DaGAN VoxCeleb

1

Hello, I see, that you've released depth face model trained on voxceleb2 Does it show better results, than your previous depth checkpoints? Can I use it with SPADE or standart...

Overfitting

5

Hello, can you please suggest me any tricks to prevent syncnet overfitting? Decreasing learning rate doesn't help... 

Code

2

Hello) How soon do you plan to release the code? Looking forward for it, thanks for your work

Multis-speaker identity degradation

4

Hi! I've encountered a problem I have multi speaker dataset. If I train a separate model for speaker (single speaker model) - prosody, speed, intonations, timbre, identity are good (for...

Hello! I see that FP16 training is default for this repo. Had someone found out, how does FP32 or mixed precision training behave comparing to FP16? Does full precision model...

Stress a vowel manually

5

Hello, thanks for your impressive work Is it possible to make a vowel stressed with some symbol? For example + or ' So we can manipulate on what phoneme will...

Nice work! What's the pipeline performance? For example, how much does it take to process 60 seconds of fullhd video?

Stress a vowel manually

5

Hello Thanks for your exciting work! Can you please tell me, is there a possibility to stress a vowel manually with you phonemizer? For example: alibab'a / alib'aba, з'амок /...

ONNX converting issues

12

Hello, I faced these errors while converting to onnx TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow...

[Bug] Training XTTSv2 from Coqui TTS leads to weird training lags with DDP

1

### Describe the bug Hello, training XTTSv2 from Coqui TTS leads to weird training lags with using DDP x6 RTX a6000 and 512GB RAM Here is monitoring GPU load graph....