MIDI Training Platform
- Piano, drums, sax & theory
- Rhythm & practice modes
- Real engraved scores, in your browser
We build 100% open-source AI software for music education and creation.
From MIDI practice and notation editing to OMR and AI audiobooks—tools that run on your own hardware, no cloud required.
Explore Our ToolsOur Projects
MEI Notation Editor
- Keyboard note input with undo
- Block arranging across staves
- Repeats, voltas & playback
AI Audiobook Pipeline
- ln-vox — Self-hosted multi-voice AI creator
- Narrative, lecture & scenario modes
- ln-reader — Android player with text sync
Rehearsal App
- Your script as a conversation
- Mutes your lines, plays every cue
- 100% on-device, no tracking
- Gemma-based vision-language model
- Reads fingerings, ornaments & dynamics
- Single & multi-voice scores
- Vision detector + graph neural network
- PDF or PNG to Humdrum **kern
- End-to-end score transcription
How It Works & Our Technology
Our mission is to democratize music education and technology by developing advanced, 100% open-source artificial intelligence software. We build innovative tools, ranging from Optical Music Recognition (OMR) engines and a keyboard-driven MEI notation editor to MIDI practice tools for piano, drums and sax, self-hosted AI pipelines for multi-voice audiobooks, and on-device rehearsal apps for performers. Our focus right now is squarely on the software itself: everything we make runs on your own hardware—in the browser, on your desktop, or on a single consumer GPU—with no accounts, no subscriptions, and no cloud required.
Open-source
models & tools
models & tools
Musicians
& Developers
& Developers
Lessons
& Creations
& Creations
Accessible for
everyone
Runs on your
own hardware
Cutting-edge AI
technology