https://github.com/magenta/magenta-demos

Demonstrations of Magenta Models
https://github.com/magenta/magenta-demos

Last synced: 20 days ago
JSON representation

Demonstrations of Magenta Models

• Host: GitHub
• URL: https://github.com/magenta/magenta-demos
• Owner: magenta
• License: apache-2.0
• Created: 2017-07-07T21:32:07.000Z (over 7 years ago)
• Default Branch: main
• Last Pushed: 2024-03-14T19:48:55.000Z (8 months ago)
• Last Synced: 2024-08-05T19:25:03.631Z (3 months ago)
• Language: Jupyter Notebook
• Size: 52.4 MB
• Stars: 1,301
• Watchers: 60
• Forks: 417
• Open Issues: 42
• Metadata Files:
  • Readme: README.md
  • License: LICENSE
  • Authors: AUTHORS

Awesome Lists containing this project

README

        
# Magenta Demos

This repository contains various demonstrations of [Magenta](https://magenta.tensorflow.org) models.

For the Magenta library and models, see our [main repo](https://github.com/tensorflow/magenta).

* [**AI Duet**](/ai-duet): A browser-based piano that responds to,

demonstrating a Magenta note sequence generation model. Built by

[Yotam Mann](https://github.com/tambien) with friends on the Magenta and

Creative Lab teams at Google. A hosted version is available as an

[AI Experiment](https://aiexperiments.withgoogle.com/ai-duet).

* [**AI Jam (Ableton)**](/ai-jam-ableton): A demonstration of several

Magenta note sequence generation models originally developed as a demonstration

for NIPS 2016. Requires an installation of Ableton Live and MaxForLive. You can

watch a video of this demo in action on our

[blog](https://magenta.tensorflow.org/2016/12/16/nips-demo/).

* [**AI Jam (JavaScript)**](/ai-jam-js): A demonstration that combines

much of the functionality of the AI Jam demo with the browser-based interface

of the AI Duet demo. Can be set up much more easily than the Ableton version.

* [**Jupyter Notebooks**](/jupyter-notebooks): Jupyter notebooks for all Magenta projects.

* [**NSynth**](/nsynth): Two demonstrations of our neural audio synthesis

method.

* [**Performance RNN Browser Demo**](/performance_rnn): Port of the [Performance RNN](https://magenta.tensorflow.org/performance-rnn) model to the [TensorFlow.js](https://js.tensorflow.org) environment.

* [**Sketch-RNN**](/sketch-rnn-js): A JavaScript implementation of our Sketch-RNN algorithm, along with 100 pre-trained models.

* [**Piano Genie**](/piano-genie-js): The user interface for Piano Genie, a demo that allows you to control a full piano via an intelligent 8-button interface.