Analysis updated 2026-05-18
Remove vocals from a song to create a karaoke track.
Mute or solo individual instruments to practice playing along with a song.
Split a song into stems using different separation models depending on speed and quality needs.
Process audio entirely offline without uploading files anywhere.
| mimrock/musichammer | aimer-zero/redforge-ai | arthuryangx/nano-notebooklm | |
|---|---|---|---|
| Stars | 41 | 41 | 41 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 3/5 | 4/5 | 2/5 |
| Audience | general | developer | researcher |
Figures from each repo's GitHub metadata at analysis time.
Downloads a large audio engine (1.5-3.5 GB) on first launch, unsigned Windows installer triggers a SmartScreen warning.
MusicHammer is a desktop app that splits a song into its separate parts, such as guitar, vocals, drums, and bass, so you can practice along with the song while muting or soloing any of those parts, or strip out the vocals entirely to make a karaoke track. Everything happens on your own computer. Your audio files never get uploaded anywhere and the app does not send any data over the internet. Installers are available for Windows 10 and 11, and for Linux as an AppImage or a .deb package, from the project's Releases page. The first time you open the app, it asks permission before downloading its audio processing engine, which is about 1.5 gigabytes, or 3.5 gigabytes if you have an NVIDIA graphics card. Individual song-splitting models are downloaded separately as needed, and each one's license is shown inside the app before you use it. On Windows, the installer is not digitally signed, so Windows SmartScreen will show a warning the first time you run it. Several different splitting models are supported, and they trade off speed, quality, and how many parts they can separate. Some install automatically the first time you use them, while others require manually clicking a download button. Performance also varies a lot depending on whether you are running on a regular processor or an NVIDIA graphics card, with GPU processing being noticeably faster for the slower models. The README describes the software as being in early development and warns that it may not work reliably and comes with no guarantees, so it should be used at your own risk. It is released under the MIT license.
A local desktop app that splits songs into separate tracks like vocals, drums, and guitar so you can mute, solo, or remove parts for practice or karaoke.
Mainly Python. The stack also includes Python, Demucs, BS-RoFormer.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly general.
This repo across BitVibe Labs
Verify against the repo before relying on details.