-
i want to run the sid pretrain model, but i got an error like this:
`generate_class.py: error: argument --task: invalid choice: 'speecht5' (choose from 'masked_lm', 'cross_lingual_lm', 'translation',…
-
Hi.
I tried pre-trained RawNet on VoxCeleb and it's works better than ECAPA-TDNN pre-trained on VoxCeleb. But when i'm switching to another dataset on English(without training) or other languages it'…
-
I find in the system description, you say adding one music file and one noise file. but in code, it seems adding one music file and 3-8 speech file.
-
This project is great. In paper, i see that Ecapa-TDNN model have two config (C=512, C=1024). Your pretrained model you uploaded has C = 1024. Could you share the pretrained model with C=512.
Thank y…
-
Hi!
I'm having trouble understanding ECAPA-TDNN architecture.
![image](https://user-images.githubusercontent.com/116825679/208719204-bef878bc-4b2e-42dc-9b99-01d128623e0c.png)
To be specific,…
-
-
## General Question
尝试用vc2中提供的预训练模型跑了下voice cloning,但音色和参考人音色差异还蛮大呢,请问下,添加参考人音频重训speaker encoder是否会有改善呢;如果重训的话,音频采样率都是16k嘛?
另外,speaker encoder采样率是16k,声码器和合成器采样率是24k,这里采样率不匹配是不影响vc效果的吧
-
Hello:
When I train ecapa-tdnn with voxceleb2, I often have the following problems. I want to ask how to solve these problems? Hope to get a reply, thank you very much!
![image](https://user-images.…
-
# Intro
Hi, I followed AMI recipe and successfully trained a model (actually tuned p value which is 0.007). I got very similar dev/eval DER scores with reported ones. Then I tried to do inference wit…
-
I tried to run AMI example: ''python experiment.py hparams/ecapa_tdnn.yaml'', but reported an error
![微信图片_20210519195057](https://user-images.githubusercontent.com/56509789/118807988-8db91e00-b8db-1…