Comments (22)
Hi,
Sorry about that
it seems the file :
"c:\sd.webui\webui\extensions\sd-wav2lip-uhq\scripts\wav2lip\face_detection\detection\sfd\s3fd.pth" does not exist, so it try to download one on the net ...
are you sure that you place this file in the good folder?
let me know :)
from sd-wav2lip-uhq.
Well it "sort of" existed. the downloaded file was named s3fd-619a316812.pth, so I renamed it to s3fd.pth and it seems to be working now.
Oh BTW the resize slider in your GUI isn't working. . .
Thanks for your help and your quick attention to my issue.
from sd-wav2lip-uhq.
True for the slider, it doesn't work as expected...
can you tell me what is your youtube channel? i'll follow for see results :)
Thanks for report and see you :)
from sd-wav2lip-uhq.
But wait, there's more. . .It failed again while it was generating, different thing this time. . . Hopefully we can get this sorted out.
https://www.youtube.com/channel/UCypaKOXWzzTxDvr3jWlfD6g I have been a huge fan of SadTalker, and I intend to compare Wav2Lip with Sad Talker. Hopefully I can get this working so that I can do so. Here is an earlier video that I did about SadTalker, and I actually mentioned the early work on Wav2Lip in the video as a foundation for SadTalker.
from sd-wav2lip-uhq.
please can you update sd-wav2lip-uhq and restart the UI?
please, let me know if this error come again
from sd-wav2lip-uhq.
Using the current commit, 6967d1a in A1111, and I did a git pull on the folder and it said it was up to date. It gave me the same error after it generated the first frame. Seems like it is having trouble saving the image. . .
from sd-wav2lip-uhq.
And did you restart the UI?
it's important, because sometime it keep parameters in memory...
I ever had the same error on different projet, like deforum and my only solution was to completly restart UI ... i'm not sure it's especially due to wav2lip..
Let me know
from sd-wav2lip-uhq.
Yep, restarted the Server and the ui
from sd-wav2lip-uhq.
ok and did you do something else in the UI before play with wav2lip?
from sd-wav2lip-uhq.
Yes, I made a video on SadTalker, But I reloaded after that
from sd-wav2lip-uhq.
that come from the option "grid_save" in options , let me check something.... i'm back quickly
from sd-wav2lip-uhq.
just for test if that came from here, can you go to :
Settings => Saving images/grids => and uncheck "Always save all generated image grids" please?
and retry
Sorry about those inconvenient...
from sd-wav2lip-uhq.
That seems to have solved it, at least as far as we can tell, but it still has lots of frames to process. I will let you know later. . .
from sd-wav2lip-uhq.
Ok i see...
do you have the same options in the "paths for savings settings"?
from sd-wav2lip-uhq.
Nope, I generate tons of images and they all go to a similar directory structure on a different drive. If I used my main drive it would be full in a week HaHA
from sd-wav2lip-uhq.
I just looked and the images are being stored in the correct directories within the sd-wav2lip-uhq folder. I'm guessing that this will all work out!
from sd-wav2lip-uhq.
ok i find a way to ignore this issue, you can re-check "Always save all generated image grids"
and after updating wav2lip-uhq the issue will not appear
from sd-wav2lip-uhq.
Well I've got good news and bad news for you:
It generated a video, and the lip syncing seems to be working well, but it attached themask to the frame and rendered them together. (see picture) github won't let me upload the video because it is too large, but the video has both the mainframe and the mask side by side.
Kind of useless if you ask me. . .
Hope that you can get this fixed.
from sd-wav2lip-uhq.
I went ahead and updated to the new commit that you wrote and reclicked the save grid thing. I'm regenerating this and it looks like it is going to work! I will let you know in the morning.
from sd-wav2lip-uhq.
Ahhhh cool !!
Let me know :)
from sd-wav2lip-uhq.
Yes, that solved the problem. Now I can get down to work! Thanks again for all your help.
from sd-wav2lip-uhq.
Perfect !! thank you too for your time !!
I close the issue :)
from sd-wav2lip-uhq.
Related Issues (20)
- conda Conflict HOT 7
- wav2lip_gan model error
- RuntimeError: Torch is not able to use GPU HOT 11
- 希望作者能出一个版本 脱离sd的,只想要这一个功能 sd 又特别笨重 HOT 7
- fatal: unable to access 'https://github.com/numz/sd-wav2lip-uhq/': Recv failure: Connection was reset ' HOT 3
- RuntimeError: Couldn't load custom C++ ops. This can happen if your PyTorch and torchvision versions are incompatible, or if you had errors while compiling torchvision from source. HOT 2
- Time out uploading data to server
- The Fix for Torch and GPU problem HOT 32
- Is it possible to not use/install bark at all???? It's causing problems and errors: (ValueError: need at least one array to concatenate) HOT 2
- Can I install an older version that does not have BARK? HOT 3
- How much time do you need to lip sync a 10 sec or 1 minute video? HOT 2
- RuntimeError: unexpected EOF, expected 45345 more bytes. The file might be corrupted.
- The video went blurry after the face swap
- issue withe the colab version
- RuntimeError: Couldn't load custom C++ ops. HOT 6
- 期待支持 wav2lip-288
- Suggestion to Extend Support for Comfy UI HOT 2
- moov atom not found / float division by zero HOT 2
- Stable Diffusion WEBUI stop working after installing "sd-wav2lip-uhq". CUDA TorchAudio issue. On clean- new virtual env. HOT 7
- Number of frames available for inference: 936 then no run HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from sd-wav2lip-uhq.