# jisungk/deepjazz

Deep learning driven jazz generation using Keras & Theano!

Repository: https://github.com/jisungk/deepjazz
Canonical: https://ross.abutalabs.com/products/deepjazz
Homepage: http://deepjazz.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: deep-learning, machine-learning, music, jazz, keras, theano, neural-networks, rnn, lstm
Last push: 2026-03-19T03:10:31+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 73, release rhythm 35, longevity 100
- inputs: {"age_days": 3806, "days_push": 167, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2890, forks 440 (observed 2026-08-28T04:07:28.399094+00:00)

## What it is
deepjazz is a Python application that uses a two-layer LSTM built with Keras and Theano to generate jazz music from a given MIDI file. It was built as a hackathon project and is no longer actively developed.

## Use cases
- generate jazz music with deep learning
- train an LSTM on a MIDI file to compose music
- learn how to build a music generation model with Keras
- create AI-generated jazz melodies
- experiment with sequence models for music
- use deep learning for algorithmic composition

## When to choose
- you want a simple, well-known reference example of LSTM-based music generation
- you are studying how deep learning can be applied to MIDI music composition
- you need educational material for a course on AI and music

## When to avoid
- you need a maintained or production-ready music generation tool
- you want to use modern frameworks like TensorFlow 2 or PyTorch
- you need native support for arbitrary MIDI files without manual preprocessing
- you require GPU support beyond NVIDIA CUDA or current library versions

## Facets
- artifact type: application
- maturity: abandoned
- function: deep-learning, machine-learning, audio-processing, data-generation
- domain: deep-learning, machine-learning
- platform: python, cli
- tags: jazz, midi, lstm, keras, theano, music-generation, generative-music, rnn, audio, gpu

## Member repositories
- jisungk/deepjazz (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:28.399094+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:35:20.431714+00:00, confidence not recorded.
  - readme: https://github.com/jisungk/deepjazz (fetched 2026-08-28T04:07:28.399094+00:00, sha 1b68fb990f8a)
  - homepage: http://deepjazz.io (fetched 2026-08-29T09:50:29.783226+00:00, sha bd4bbcda9538)
- Data as of 2026-08-30T08:39:29.467469+00:00.