A Novel Dataset for the Identification of Computer Generated Melodies in the CSMT Challenge

Shengchen Li*, Yinji Jing, György Fazekas

*Corresponding author for this work

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review


This paper introduces a novel dataset for the identification of computer generated melodies as used in the data challenge organised by the Conference on Sound and Music Technology (CSMT). The CSMT data challenge requires participants to identify whether a given piece of melody is generated by computer or is composed by human. The dataset consists of two parts: a development dataset and an evaluation dataset. The development dataset contains only computer generated melodies whereas the evaluation dataset contain both computer generated melodies and human composed melodies. The aim of the dataset is to facilitate the develpment and assessment of methods to identified computer generated melodies and facilitate the creation of generative music systems.

Original languageEnglish
Title of host publicationProceedings of the 8th Conference on Sound and Music Technology - Selected Papers from CSMT
EditorsXi Shao, Kun Qian, Li Zhou, Xin Wang, Ziping Zhao
PublisherSpringer Science and Business Media Deutschland GmbH
Number of pages10
ISBN (Print)9789811616488
Publication statusPublished - 2021
Event8th Conference on Sound and Music Technology, CSMT 2020 - Taiyuan, China
Duration: 5 Nov 20208 Nov 2020

Publication series

NameLecture Notes in Electrical Engineering
Volume761 LNEE
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119


Conference8th Conference on Sound and Music Technology, CSMT 2020


  • Computer generated melody identification
  • Dataset
  • Melody clustering

Cite this