Making music using code isn’t a new thing but machine learning gives us a whole new way to go about it. We played the computer tons of examples of melodies and over time it starts to learn relationships between notes and timings. So when you play a few notes the neural network will respond with a short melody, interpreting whatever you give it. It’s like a musical game of call and response.

This is not an official Google product.

CREDITS

Built by Yotam Mann and folks on the Magenta and Creative Lab teams at Google.

OVERVIEW

A.I. Duet is composed of two parts, the front-end which is in the 'static' folder and the back-end which is in the 'server' folder. The front-end client creates short midi files using the user's input which is sent to the server. The server takes that midi input and continues it using Magenta which is then returned back to the client.

FRONT END

To build the client-side javascript, first install node and webpack. Then you can install of the dependencies of the project by typing the following in the terminal:

cd static
npm install

Then build all of the files

webpack -p

BACK END

The back-end consists of a Flask server, TensorFlow and Magenta.

First install TensorFlow.

To install the rest of the back-end requirements (make sure you have python and pip installed):

cd server
pip install -r requirements.txt

Then run the server:

python server.py

You can now play with A.I. Duet at localhost:8080.

MIDI SUPPORT

The A.I. Duet supports MIDI keyboard input using Web Midi API and the WebMIDI library. Make sure you refresh the page after plugging in a keyboard to make sure it connects correctly.

PIANO KEYBOARD

The piano can also be controlled from your computer keyboard thanks to Audiokeys. The center row of the keyboard is the white keys.

PIANO AUDIO

Multisampled piano from Salamander Grand Piano V3 by Alexander Holm (Creative Commons Attribution 3.0).

LICENSE

Copyright 2016 Google Inc.

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.