Skip to content

Video-Retalking.md

Latest commit

 

History

History
88 lines (70 loc) · 3.47 KB

Video-Retalking.md

File metadata and controls

88 lines (70 loc) · 3.47 KB
Raw

Launching After Installation

  • set PYTHONPATH=%PYTHONPATH%;C:\Users\bryan\videoretalkingENV\lib\site-packages
  • cd C:\Users\bryan\video-retalking\ && C:\Users\bryan\videoretalkingENV\Scripts\activate.bat
  • python webui.py

Note: Firefly images are high-resolution so you may need to downsize before using the software. Then after the video is created use your editor or ffmpeg to rework the quality.

Notes on Installation (follow instruction of Repo)

TOP ISSUES ON INSTALLATION

  • Please follow the instructions from https://pytorch.org/get-started/previous-versions/
  • The biggest issues are : torch==1.9.0+cu111 torchvision==0.10.0+cu111
  • set PYTHONPATH=%PYTHONPATH%;C:\Users\bryan\videoretalkingENV\lib\site-packages
  • downgrading to numpy==1.23.1

This installation command only works on CUDA 11.1

  • pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 -f https://download.pytorch.org/whl/torch_stable.html
  • Successfully installed torch-1.9.0+cu111 torchvision-0.10.0+cu111
  • pip show torch
  • pip show torchvision
  • pip show numpy
  • pip install -r requirements.txt

Paths Are Important (highly recommend ENVs)

  • set PYTHONPATH=%PYTHONPATH%;C:\Users\bryan\videoretalkingENV\lib\site-packages
  • import sys
  • print(sys.path)

Checking Python Versions & Launching WebUI

  • python --version
  • python3 --version
  • python webui.py
  • python3 webui.py

Video Cleanup & wav to mp4

  • ffmpeg -i input.mp3 output.wav
  • ffmpeg -i input.mp3 -b:a 16k -ac 1 output.wav
  • ffmpeg -i c:\users\bryan\desktop\PixelWorks.mp4 -vf "deblock,hqdn3d,minterpolate='fps=30'" c:\users\bryan\desktop\video_fixed.mp4
  • ffmpeg -vcodec mpeg4 -b:v 7561k -qscale:v 2 -acodec aac -ac 2 -async 1 -strict experimental c:\users\bryan\desktop\video_fixed.mp4 -threads 0 -i c:\users\bryan\desktop\PixelWorks.mp4
  • ffmpeg -i c:\users\bryan\desktop\PixelWorks.mp4 -af "pan=stereo|c0=c0|c1=c0" c:\users\bryan\desktop\video_stereo.mp4
  • .wav and .mp4 >>> name the same.

IF HAVING ISSUES CHECK OUT THIS ISSUE AND DOWNGRADE NUMPY (CRITICAL)

Troubleshooting (Using CLI)

  • python inference.py --face temp/video\1.mp4 --audio temp/audio\1.wav --outfile results/NEW_1.mp4
  • (videoretalkingENV) C:\Users\bryan\video-retalking>python inference.py --face temp/video\3.mp4 --audio temp/audio\1-0-100.wav --outfile results/pixel_0.mp4
  • (videoretalkingENV) C:\Users\bryan\video-retalking>python inference.py --face temp/video\1.mp4 --audio temp/audio\1.wav --outfile results/NEW_1.mp4

Batch File for Your Desktop

  • @echo off

  • echo https://github.com/OpenTalker/video-retalking

  • echo Python311

  • echo Python311-32

  • echo Python38

  • echo Python39-32

  • echo Opening Explorer...

  • start sndvol

  • start explorer "C:\Users\bryan\video-retalking"

  • timeout 5

  • echo Opening Task Manager / Resource Monitor...

  • start taskmgr

  • start resmon

  • timeout 5

  • echo Activating Virtual Environment...

  • call "C:\Users\bryan\videoretalkingENV\Scripts\activate.bat"

  • echo Starting Video-Retalking Application...

  • cd "C:\Users\bryan\video-retalking"

  • "C:\Users\bryan\videoretalkingENV\Scripts\python.exe" "C:\Users\bryan\video-retalking\webui.py"

  • exit