Git Product home page Git Product logo

dl4mir's Introduction

dl4mir: A Tutorial on Deep Learning for MIR

by Keunwoo Choi ([email protected])

This is a repo for my tutorial paper; A Tutorial on Deep Learning for Music Information Retrieval.

Tutorials

  1. Example 1: Pitch detector with a dense layer
  2. Example 2: Chord recogniser with a convnet
  3. Example 3: Setup config.json
  4. Example 4: download and preprocess
  5. Real examples with real datasets! * Example 5-1: Time-varying classification example using Jamendo dataset * Example 5-2: Time-invariant classification example using FMA dataset

Prerequisites

$ pip install -r requirements.txt
$ git clone https://github.com/keunwoochoi/kapre.git
$ cd kapre
$ python setup.py install

to install

  • Librosa, Keras, Numpy, Matplotlib, Future
  • kapre

Notes

  • Datasets is removed from Kapre and the codes are directly imported into here.

Datasets

Dataset management

  • GTZan: (30s, 10 genres, 1,000 mp3)
  • MagnaTagATune: (29s, 188 tags, 25,880 mp3) for tagging and triplet similarity
  • MusicNet: (full length 330 classicals music, note-wise annotations)
  • FMA: small/medium/large/full collections, up to 100+K songs from free music archieve, for genre classification. With genre hierarchy, pre-computed features, splits, etc.
  • Jamendo: 61/16/24 songs for vocal activity detection

Some links

  • Repo
  • Slides
    • Deep Neural Networks in MIR: A tutorial focusing on feature learning, beat/rhythm analysis, structure analysis. Also a nice literature overview including publications by year, conference, task, network types, input representations, frame work, etc. By Meinard Muller et al.
    • DL in music informatics: ISMIR 2014 tutorial.
  • Documents, books
    • Deep learning book: The first deep learning textbook. by Ian Goodfellow and Yoshua Bengio and Aaron Courville.
  • Online

Cite?

@article{choi2017tutorial,
  title={A Tutorial on Deep Learning for Music Information Retrieval},
  author={Choi, Keunwoo and Fazekas, Gy{\"o}rgy and Cho, Kyunghyun and Sandler, Mark},
  journal={arXiv:1709.04396},
  year={2017}
}

Or visit the paper page on Google scholar for potential updates.

dl4mir's People

Contributors

keunwoochoi avatar lvaleriu avatar

Watchers

James Cloos avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.