Skip to content

UVR5 UI v1.6.0
Compare
Choose a tag to compare
@Eddycrack864 Eddycrack864 released this 27 Nov 21:05
· 19 commits to main since this release
v1.6.0
5a74429
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Important Installation Guidelines

Before you install UVR5 UI

Keep these instructions in mind:

  • File Location: Place the UVR5 UI files in a folder on your C drive (e.g. /Downloads folder).
  • Access Control: Avoid placing UVR5 UI in folders that require Admin privileges.
  • Path Requirements: Ensure the installation path is free of spaces or special characters.

Installation Steps

  1. Download: Clone the repository or download the source code from this link.
  2. Run the Installer: Inside the folder run the UVR5-UI-installer.bat (Windows) or UVR5-UI-installer.sh (Linux) script.
  3. Start UVR5 UI: Launch UVR5 UI by running the run-UVR5-UI.bat (Windows) or run-UVR5-UI.sh (Linux) script.

What's new in this release?

  • New models added:
    • MelBand Roformer Kim | Inst V1 by Unwa
    • MelBand Roformer Kim | Inst V2 by Unwa
    • MelBand Roformer Kim | InstVoc Duality V1 by Unwa
    • MelBand Roformer Kim | InstVoc Duality V2 by Unwa
  • Inference speed improved by 80%
Audio length Duration of inference before Duration of inference after
269s 209s 41s
321s 250s 49s
221s 175s 36s

This is possible thanks to the use of Pytorch Autocast and the new code improvements. Tested on Tesla T4 btw.

  • New supported audio formats:
    • ogg
    • opus
    • m4a
    • aiff
    • ac3
  • Progress bar added
  • Stop renaming input audio files (The name of the original audio is now preserved)
  • Stop using CLI (use python function instead)
  • Code Improvements
  • New features have been added to the interface (Batch Size, Amplification Threshold, Normalization Threshold and more!)

Full Changelog: v1.5.2...v1.6.0

Assets 2
Loading