AuTGeLy: Automatic Title Generator based on Song Lyrics Extractions

Diego Vallejo-Huanga, Esteban Carrera, Jonathan Manay

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

Abstract

Natural Language Processing (NLP) is one of the most popular fields of Artificial Intelligence and its applications are diverse. This article presents a recommender system, through a web application, that extracts the lyrics of a song entered in audio format and uses NLP techniques to process the textual corpus of the lyrics and recommend a title for the song. The level of effectiveness of the system was analyzed, measuring the percentage of similarity between the ground truth, which is the original title of the song, and our recommendation. For the experiments, a dataset of 30 songs was used divided into three taxonomies, small, medium, and large, that change according to the length of tokens of the original title. The results show an accuracy of 70% for small titles and 20% for medium and long titles. It is also shown that the web tool is enabled to formulate control of lexical content in songs since it does not use the original title of the song as input.

Original languageEnglish
Title of host publicationProceedings - 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021
EditorsM. Arif Wani, Ishwar K. Sethi, Weisong Shi, Guangzhi Qu, Daniela Stan Raicu, Ruoming Jin
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1588-1593
Number of pages6
ISBN (Electronic)9781665443371
DOIs
StatePublished - 2021
Event20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021 - Virtual, Online, United States
Duration: 13 Dec 202116 Dec 2021

Publication series

NameProceedings - 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021

Conference

Conference20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021
Country/TerritoryUnited States
CityVirtual, Online
Period13/12/2116/12/21

Bibliographical note

Funding Information:
This work was supported by IDEIAGEOCA Research Group of Universidad Politécnica Salesiana in Quito, Ecuador.

Publisher Copyright:
© 2021 IEEE.

Keywords

  • Content Analysis
  • LDA
  • Natural Language Processing
  • Song Lyrics Extraction.
  • Speech Recognition

Fingerprint

Dive into the research topics of 'AuTGeLy: Automatic Title Generator based on Song Lyrics Extractions'. Together they form a unique fingerprint.

Cite this