User specific training of a music search engine

David Little*, David Raffensperger, Bryan A Pardo

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Query-by-Humming (QBH) systems transcribe a sung or hummed query and search for related musical themes in a database, returning the most similar themes as a play list. A major obstacle to effective QBH is variation between user queries and the melodic targets used as database search keys. Since it is not possible to predict all individual singer profiles before system deployment, a robust QBH system should be able to adapt to different singers after deployment. Currently deployed systems do not have this capability. We describe a new QBH system that learns from user provided feedback on the search results, letting the system improve while deployed, after only a few queries. This is made possible by a trainable note segmentation system, an easily parameterized singer error model and a straight-forward genetic algorithm. Results show significant improvement in performance given only ten example queries from a particular user.

Original languageEnglish (US)
Title of host publicationMachine Learning for Multimodal Interaction - 4th International Workshop, MLMI 2007, Revised Selected Papers
Pages72-83
Number of pages12
DOIs
StatePublished - Mar 10 2008
Event4th International Workshop on Machine Learning for Multimodal Interaction, MLMI 2007 - Brno, Czech Republic
Duration: Jun 28 2007Jun 30 2007

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4892 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other4th International Workshop on Machine Learning for Multimodal Interaction, MLMI 2007
Country/TerritoryCzech Republic
CityBrno
Period6/28/076/30/07

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'User specific training of a music search engine'. Together they form a unique fingerprint.

Cite this