Unleashing your musical creativity with Magenta.js
summary
Magenta.js is a browser library that runs TensorFlow.js models to generate and transform music and images: MIDI-based generators that produce navigable melodies, a chorale harmonizer trained on 306 Bach chorales, an eight‑button Piano Genie that maps to 88-note output, QuickDraw stroke completion, and style transfer all in a few lines of code. Demo tools include five Ableton plugins (Magenta Studio), a MusicVAE sampler, and remixable demos hosted on glitch remixable hosting. Installations include infinite GAN portraits and one‑off generated skull paintings that vanish.
AI-written summary · about
chapters (31)
- 0:00Intro
- 0:38Why Magenta
- 3:31Making things is good
- 6:24Rules
- 8:26Platform
- 10:05Art
- 13:30Serious
- 14:37What is Machine Learning
- 15:13Functions
- 16:05Models
- 16:53Why is it so good
- 17:35Compute power
- 18:51generative models
- 22:12generative network
- 23:49algorithms
- 24:10Automation
- 24:55Incredible Spinner
- 26:39Bach Doodle
- 28:13Companion App
- 30:14Piano Genie
- 30:55Sample
- 32:05Style Transfer
- 32:49Drawing
- 33:57Magentajs
- 34:23Code Breakdown
- 36:49Code Should Never Hold Back Art
- 37:39The Medium is the Message
- 38:17Memories of a passerby
- 39:04Random Skulls
- 41:25Building serious things
- 42:13Questions
related talks
28mMusic and AI in the Browser with TensorFlow js and Magenta js
46mMusic Generation with Magenta: Using Machine Learning in
41mMusic and Machine Learning
24mCreating an AI Musician with JavaScript
40mLiveJS: Making Music with JavaScript
14mMusical Creativity in the Browser
39mMachine Learning Magic for Your JavaScript Application
50mJavaScript for Artists