muscriptor/muscriptor

Health v2 · maintenance only

Flags: young

Adoption not part of the score

1263 stars · 149 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

MuScriptor is a multi-instrument music transcription model by Kyutai and Mirelo that converts audio recordings into MIDI and sheet music. It is distributed as a Python package with a CLI and a self-hostable web UI, and is claimed to be the most accurate open-source transcription model.

Use cases

• transcribe a song recording to MIDI
• convert audio to sheet music notation
• transcribe multi-instrument music files
• self-host a music transcription web app
• run audio-to-MIDI transcription from the command line

When to choose

• you need open-source audio-to-MIDI transcription with high accuracy
• you want MIDI plus engraved sheet music output from recordings
• you want a self-hosted transcription service or CLI

When to avoid

• you need a permissively licensed model for commercial use (weights are CC BY-NC 4.0)
• you need real-time low-latency transcription
• you cannot run GPU inference locally and don't want to use the hosted service

Facets

library · maturity active

machine-learning audio-processing llm-inference machine-learning deep-learning media python windows cli music-transcription audio-to-midi sheet-music transformer music-information-retrieval self-hosted audio linux macos web-server gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="muscriptor/muscriptor")