magenta/mt3
MT3: Multi-Task Multitrack Music Transcription observed · 2026-08-28
Health v2 · maintenance only
73/100
- Activity 91
- Release rhythm 35
- Longevity 100
Flags: no_releases
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: n/a
- age_days: 1764
- days_rel: n/a
- days_push: 55
- n_releases_24m: 0
Adoption not part of the score
1744 stars · 223 forks observed · 2026-08-28
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded
MT3 is a multi-instrument automatic music transcription model built on the T5X framework, converting audio recordings into multitrack MIDI. It provides pretrained checkpoints for piano and multi-instrument transcription, primarily usable via a Colab notebook.
Use cases
- transcribe piano recordings to midi
- convert multi-instrument audio into midi files
- automatic music transcription from audio
- extract notes from a song recording
- transcribe audio files with pretrained model in colab
When to choose
- you need audio-to-MIDI transcription of piano or multiple instruments
- you want to use pretrained transcription checkpoints without training your own
When to avoid
- you want to train or fine-tune a transcription model yourself, since training is not easily supported
- you need a production-ready application or API rather than a research model
Facets
library · maturity maintenance
machine-learning deep-learning audio-processing nlp machine-learning artificial-intelligence python cloud music-transcription audio-to-midi transformers t5x multitrack ismir audio gpu
1 source
- readme: https://github.com/magenta/mt3 · fetched 2026-08-28 · d1fb927d6887
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| magenta/mt3 | main | 73 |
For agents
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem