PhD in Computer Science

2015 - 2018

My PhD in one sentence (Other PhD description in one sentence)
  • Building playlist is hard.
Aim
  • Provide new tools for high-level music analysis
  • Enhance the supervised classification of musical tags in big unbalanced musical databases.
Fields
  • Machine learning
  • Artificial intelligence
  • Digital audio signal processing
  • Music Information Retrieval
  • Musical recommender systems
  • Music autotagging
Supervisors
Affiliations
Conference presence
Reviewer for

Articles

Journal

Detecting temporal changes in acoustic scenes: The variable benefit of selective attention.

Laurent Demany, Yann Bayle, Emilie Puginier and Catherine Semal
Hearing Research (IF: 2.9)
Volume 353, September 2017, Pages 17–25
Pdf Icon
Conference paper

Kara1k: a karaoke dataset for cover song identification and singing voice analysis

Yann Bayle, Ladislav Maršík, Martin Rusek, Matthias Robine, Pierre Hanna, Kateřina Slaninová, Jan Martinovič and Jaroslav Pokorný
19th IEEE International Symposium on Multimedia (ISM)
Taichung, Taiwan
11-13 December 2017
Pages 1-8
Acceptance rate: 21.82%
Honorable mention for being in the Top 6 papers
Source code
Pdf Icon

SATIN: A Persistent Musical Database for Music Information Retrieval.

Yann Bayle, Pierre Hanna and Matthias Robine
15th International Workshop on Content-Based Multimedia Indexing (CBMI)
Florence, Italy
19-21 June 2017
Pages 1-5
Source code
Pdf Icon

Large-Scale classification of music tracks according to the presence of singing voices.

Yann Bayle, Pierre Hanna and Matthias Robine
10ème conférence des Journées d’Informatique Musicale (JIM).
Albi, France
31 March 2016
Pages 144-152
Pdf Icon