# googlecreativelab/aiexperiments-ai-duet

A piano that responds to you.

Repository: https://github.com/googlecreativelab/aiexperiments-ai-duet
Canonical: https://ross.abutalabs.com/products/aiexperiments-ai-duet
Homepage: https://aiexperiments.withgoogle.com/ai-duet
Language: JavaScript
License Family: other
Archived: true
Last push: 2025-08-18T19:18:03+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 37, release rhythm 35, longevity 100
- inputs: {"age_days": 3583, "days_push": 380, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1668, forks 298 (observed 2026-08-28T04:05:19.781288+00:00)

## What it is
A.I. Duet is a Google Creative Lab experiment where a neural network trained on MIDI examples responds to notes you play on a virtual piano. It consists of a JavaScript front-end that sends MIDI input to a Flask server that generates continuations using Magenta and TensorFlow.

## Use cases
- play a piano duet with an AI
- generate musical continuations from MIDI input
- experiment with machine learning music generation
- build an interactive music web app with Magenta
- learn how neural networks model note timing
- demo AI music response in the browser

## When to choose
- you want a reference example of Magenta/TensorFlow music generation
- you need an interactive browser-based music experiment to fork
- you're studying how front-end MIDI input can feed a Python ML backend

## When to avoid
- you need a maintained, production-ready music AI service
- you can't use Python 2.7 or outdated dependencies
- you need support, bug fixes, or up-to-date security patches

## Facets
- artifact type: application
- maturity: abandoned
- function: machine-learning, audio-processing, web-framework, http-server
- domain: artificial-intelligence, machine-learning, entertainment, web-development
- platform: python, cross-platform
- tags: music, midi, magenta, tensorflow, tonejs, google-creative-lab, neural-network, interactive-experiment, archived, audio, nodejs, web-server, docker

## Member repositories
- googlecreativelab/aiexperiments-ai-duet (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:19.781288+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 2026-08-30T03:42:50.108717+00:00, confidence not recorded.
  - readme: https://github.com/googlecreativelab/aiexperiments-ai-duet (fetched 2026-08-28T04:05:19.781288+00:00, sha 4de335d9a22a)
  - homepage: https://aiexperiments.withgoogle.com/ai-duet (fetched 2026-08-29T11:16:29.035915+00:00, sha 79707820d344)
- Data as of 2026-08-30T08:39:29.467469+00:00.