jisungk/deepjazz
A two-layer LSTM neural network that learns from MIDI files to generate jazz music.
Not currently ranked — collecting fresh signals.
star history
Deepjazz is a generative music project that trains a recurrent neural network on MIDI data to produce jazz compositions. It builds a two-layer LSTM architecture using Keras and Theano, processing melody data from input MIDI files to generate new musical output. The project was developed at a hackathon and outputs results that can be shared on SoundCloud.
Frequently asked
- What is jisungk/deepjazz?
- A two-layer LSTM neural network that learns from MIDI files to generate jazz music.
- Is deepjazz open source?
- Yes — jisungk/deepjazz is open source, released under the Apache-2.0 license.
- What language is deepjazz written in?
- jisungk/deepjazz is primarily written in Python.
- How popular is deepjazz?
- jisungk/deepjazz has 2.9k stars on GitHub.
- Where can I find deepjazz?
- jisungk/deepjazz is on GitHub at https://github.com/jisungk/deepjazz.