Skip to content

Repository files navigation

muski-performance-rnn

Adaptation of the Google Magenta Performance RNN Browser demo for the Muski website.

Adapted from the port of the Performance RNN model to the TensorFlow.js environment by Google Magenta.

Building

To build, execute yarn bundle. This will regenerate bundle.js, which is referenced by index.html.

To view, execute yarn run-demo, which will start a local webserver.

Credits

Original demo by Ian Simon and Sageev Oore.

Google.

Adapted for the Muski website by Eric Londaits, for IMAGINARY gGmbH.

Performance RNN was trained in TensorFlow on MIDI from piano performances from the Yamaha e-Piano Competition dataset. It was then ported to run in the browser using only Javascript in the TensorFlow.js environment. Piano samples are from Salamander Grand Piano.

License

Licensed under the Apache License, Version 2.0.

Original demo ©2017 Google. Adaptation ©2022 IMAGINARY gGmbH.

Salamander Grand Piano V3 by Alexander Holm licensed under CC-BY v3.0.

About

Adaptation of the Google Magenta Performance RNN demo for use in the Muski website

Resources

Stars

0 stars

Watchers

2 watching

Forks

Releases

Packages

Contributors

Languages