# spotify/basic-pitch

A lightweight yet powerful audio-to-MIDI converter with pitch bend detection

Repository: https://github.com/spotify/basic-pitch
Canonical: https://ross.abutalabs.com/products/basic-pitch
Homepage: https://basicpitch.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: lightweight, machine-learning, midi, music, pitch-detection, polyphonic, transcription, audio, python, typescript
Last push: 2025-11-13T14:40:46+00:00

## Health v2 (maintenance only)
Score: 46/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 52, release rhythm 8, longevity 100
- inputs: {"age_days": 1583, "days_push": 293, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 5489, forks 501 (observed 2026-08-28T04:09:19.758793+00:00)

## What it is
Basic Pitch is a lightweight neural network library from Spotify for automatic music transcription, converting audio recordings of nearly any single instrument into MIDI files with pitch bend detection. It is pip-installable in Python and npm-installable in TypeScript, and runs faster than real time on modern hardware.

## Use cases
- convert a recording of me humming into a MIDI file
- transcribe guitar or piano audio to MIDI for my DAW
- extract notes from a polyphonic instrument recording
- detect pitch bends in audio and preserve them in MIDI
- add audio-to-MIDI conversion to a Python or TypeScript app
- turn voice recordings into editable digital scores

## When to choose
- you need fast, lightweight audio-to-MIDI transcription without heavy GPU resources
- you work with single-instrument or monophonic recordings, including voice
- you want a pip/npm-installable library or a quick web demo
- you need pitch bend detection for expressive MIDI output

## When to avoid
- you need accurate transcription of multi-instrument mixes
- you require studio-grade, note-perfect transcription for professional engraving
- you need real-time streaming transcription rather than file-based conversion

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, audio-processing, nlp
- domain: machine-learning, media, audio
- platform: python, windows, cross-platform
- tags: audio-to-midi, music-transcription, pitch-detection, polyphonic, pitch-bend, midi, deep-learning, music-production, audio, macos, linux, nodejs

## Member repositories
- spotify/basic-pitch (main) score 46

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:19.758793+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:56:21.761565+00:00, confidence not recorded.
  - readme: https://github.com/spotify/basic-pitch (fetched 2026-08-28T04:09:19.758793+00:00, sha dcdb3d9a12aa)
  - homepage: https://basicpitch.io (fetched 2026-08-29T08:51:38.698113+00:00, sha a4cc9df190ce)
  - site_page: https://basicpitch.spotify.com/about (fetched 2026-08-29T08:51:38.706946+00:00, sha ca1c2d21aeeb)
- Data as of 2026-08-30T08:39:29.467469+00:00.