-
WavLMのところでバッチサイズ1のときのshapeがうまくあっていない様子
```
RuntimeError: Given groups=1, weight of size [512, 1, 10], expected input[1, 5945, 1] to have 1 channels, but got 5945 channels instead
```
修正中です
-
### Describe the bug
when trying to finetune WavLM and using DDP. there are some unused parameters. This causes the run to crash. when using --find_unused_parameters it says there are no unused par…
-
Please check whether this paper is about 'Voice Conversion' or not.
## article info.
- title: **Audio Deepfake Detection with Self-Supervised WavLM and Multi-Fusion Attentive Classifier**
- summary: …
-
Using noise scaled MAS for VITS2
Using duration discriminator for VITS2
INFO:models:Loaded checkpoint 'Data\abc\models\DUR_0.pth' (iteration 0)
ERROR:models:emb_g.weight is not in the checkpoint
I…
-
Hi, thanks for your great work! I would like to use VSim for speaker similarity evaluation. From the document, I see that I should use "wavlm_large_fintune.pth" model when "model_type=valle". I'm not …
-
跟 worker 不知道有关系么 , batch 设置为1 就会报这个错误 ,batch 高 ,占用现存太大
Traceback (most recent call last):
File "train_ms.py", line 840, in
run()
File "train_ms.py", line 361, in run
trai…
-
model | EER(mine) | EER(official)
-- | -- | --
wavlm_large_nofinetune.pth | 0.965 | 0.75
wavlm_large_finetune.pth | 0.631 | 0.431
The above result…
-
Hi, I am trying to finetune FreeVC-s model with a small dataset formatted in vctk format. I have done necessary changes in config file. I am also not using SR-based augmentation. But when I run train.…
-
안녕하세요.
굉장한 프로젝트를 오픈소스로 올려주셔서 감사합니다.
VC과 TTS를 융합하기 위해서 w2v를 뽑아내는 컨셉이 정말 재밌습니다.
흥미가 생겨서 학습을 시켜보고 있던 중에
TTV transformer 관련 질문이 생겨 이렇게 issue 올립니다.
posterior => decode 는 잘 되는 것을 확인했습니다만,
prior와 p…
-
@ductuantruong
Thank you for sharing your excellent research.
I would like to inquire about training config options.
It is known that applying speech perturbation in speaker verification per…