Open-source model that transcribes songs into separate instrument tracks and MIDI notes, including vocals.
Updated EU status checked Last shown
Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.
MuScriptor (also referred to as Muse Scriber or Muse Scriptor) is an open model for music transcription. According to a YouTube creator's review, it takes a song and splits it into separate instrument tracks, outputting MIDI notes for each, vocals included.
It is reported to be published in three sizes: a large variant (about 5.5 GB), a medium variant, and a small variant with roughly 0.1B parameters (about 412 MB). The smaller size makes local use feasible on modest hardware. The model is hosted on Hugging Face, where a demo Space is available.
The reviewer saw some transcription errors when testing the Space, so output likely needs manual cleanup. The vendor, license and release date could not be verified from official pages.
Watch the moment: the exact second top creators show it. Videos load only when you press play, via youtube-nocookie.com.
Open-source tool that transcribes a song into separate instrument tracks and MIDI notes, including vocals. The creator tried it in a Hugging Face Space and saw some errors. It comes in large (5.5 GB), medium and small (0.1B parameters, 412 MB) variants.
Closest neighbours by meaning, not just by category.
Model · Mirelo
Mirelo's Audio-to-MIDI turns songs and other audio into editable MIDI and MusicXML, with per-instrument tracks.
Model
Score-native singing voice synthesis system that turns lyrics and a MIDI melody into expressive vocals.
Model
Vision Transformer foundation model for sheet music, pre-trained with masked autoencoding on 9.7M IMSLP pages.
Model · Meta
Meta speech-to-text model described as suited to real-time streaming transcription.
Model · ElevenLabs
ElevenLabs' text-to-music model for generating songs with vocals, inpainting and genre transitions, cleared for commercial use.
Model
Open-source local AI model for generating music loops, with reference-loop stacking and MIDI conversion.
