Grant: Discovery Grants Program - Individual
University of Victoria
$20,000.00
Sound source description for music information retrieval
EF 2008-09
Résumé
Music Information Retrieval (MIR) is an emerging area of multimedia research dealing with the analysis and retrieval of music in digital form. The proposed research program is focused on improving the state-of-the-art in MIR algorithms for processing audio signals using ideas from sound source separation. In the past few years there has been steady progress in various topics of audio MIR such as genre/style/mood/artist classification, segmentation, summarization and similarity retrieval. In the majority of existing algorithms the audio signal is represented statistically without any effort to identify and characterize individual sound sources.
The main objective of the proposed research is to explore how sound source separation and music transcription techniques can be used to aid large scale MIR problems such as classification and similarity retrieval.
As an example the characteristics of the singing voice play an important role in how we characterize/perceive a piece of music. Using sound source separation techniques the singing voice part of a music mixture can be enhanced and characterized individually. This information can then be combined with other more "classic" audio features for classification or similarity retrieval. We hope that the shift to large data collections from isolated examples will suggest insights and improvements to existing source approaches and inspire new ones.
MIR is one of the most important emerging areas of multimedia information retrieval. The success of portable music players and online music stores indicates the large impact that MIR can have. In addition to commercial applications, improvements in MIR systems can lead to advances in music libraries, pedagogy and preservation of culture. The proposed research has the potential to improve existing sound source separation approaches and therefore make human/computer communication more effective.
Programme
Discovery Grants Program - Individual
Bénéficiaire
Emplacement
British Columbia, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
298270-2007
Name nom
Tzanetakis, George
Department département
Computer Science
Organization id
7
Institution établissement
University of Victoria
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2008
Competition year année de concours
2007
Award amount
20000
Program id
RGPIN
Program nane en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
Discovery grants programs
Group fr
Programmes de subventions à la découverte
Committee code
331
Committee name en
Computing and Information Sciences - B
Committee name fr
Informatique - B
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2721
Research subject group en
Information technology
Research subject group fr
Technologie de l'information
Research subject en
Multimedia systems and networks
Research subject fr
Systèmes et réseaux multimédias
Partie
2
Nb partie
5
Application title
Sound source description for music information retrieval
Keyword
NA
Application summary
Music Information Retrieval (MIR) is an emerging area of multimedia research dealing with the analysis and retrieval of music in digital form. The proposed research program is focused on improving the state-of-the-art in MIR algorithms for processing audio signals using ideas from sound source separation. In the past few years there has been steady progress in various topics of audio MIR such as genre/style/mood/artist classification, segmentation, summarization and similarity retrieval. In the majority of existing algorithms the audio signal is represented statistically without any effort to identify and characterize individual sound sources.
The main objective of the proposed research is to explore how sound source separation and music transcription techniques can be used to aid large scale MIR problems such as classification and similarity retrieval.
As an example the characteristics of the singing voice play an important role in how we characterize/perceive a piece of music. Using sound source separation techniques the singing voice part of a music mixture can be enhanced and characterized individually. This information can then be combined with other more "classic" audio features for classification or similarity retrieval. We hope that the shift to large data collections from isolated examples will suggest insights and improvements to existing source approaches and inspire new ones.
MIR is one of the most important emerging areas of multimedia information retrieval. The success of portable music players and online music stores indicates the large impact that MIR can have. In addition to commercial applications, improvements in MIR systems can lead to advances in music libraries, pedagogy and preservation of culture. The proposed research has the potential to improve existing sound source separation approaches and therefore make human/computer communication more effective.