Automatic Beatmap Generating Rhythm Game Using Music Information Retrieval with Machine Learning for Genre Detection

The study is aimed to develop an Automatic Beatmap with Genre Detection, called "Efflorescence", a mobile application which can generate a rhythm game for people who would like to improve their reflexive functions. This study also provides different music genres that will be detected durin...

Full description

Saved in:

Bibliographic Details
Published in:	2020 IEEE 12th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM) pp. 1 - 6
Main Authors:	Estolas, Elijah Alixtair L., Malimban, Agatha Faith V., Nicasio, Jeremy T., Rivera, Jyra S., Pablo, May Florence D. San, Takahashi, Toru L.
Format:	Conference Proceeding
Language:	English
Published:	IEEE 03-12-2020
Subjects:	Automatic Audio Segmentation game design Games Machine learning algorithms Music information retrieval Onset Detection Procedural Content Generation Rhythm rhythm game Software Software reliability Tempo Extraction Usability
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	The study is aimed to develop an Automatic Beatmap with Genre Detection, called "Efflorescence", a mobile application which can generate a rhythm game for people who would like to improve their reflexive functions. This study also provides different music genres that will be detected during the generation process so that users are able to distinguish different types of music among the songs they have chosen and/or uploaded to play. The researchers also aim in determining known music genres and its alternatives, and to be able to generate non-fixed beat maps to give the users a little challenge than most rhythm games produced. For the researchers to create the application, the following algorithms were used: Music Information Retrieval, Onset Detection, Tempo Detection, and Machine Learning. To prove that the application is feasible, the researchers conducted a survey among 50 respondents, all composed of FEU Institute of Technology CS and IT. The respondents rated the application average of being able to produce the result they wanted towards the game. The system can be further improved by future researchers through updating the system by putting up more functions and data required for the genre detection. It is also recommended that future researchers would apply it on different other platforms that were not and to lessen the specifications of the hardware itself. Lastly, future researchers can add more interactive features to make the game more challenging yet fun at the same time.
DOI:	10.1109/HNICEM51456.2020.9400133