Robust music identification, detection, and analysis

Mehryar Mohri, Pedro Moreno, Eugene Weinstein

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

Abstract

In previous work, we presented a new approach to music identification based on finite-state transducers and Gaussian mixture models. Here, we expand this work and study the performance of our system in the presence of noise and distortions. We also evaluate a song detection method based on a universal background model in combination with a support vector machine classifier and provide some insight into why our transducer representation allows for accurate identification even when only a short song snippet is available.

Original languageEnglish (US)
Title of host publicationProceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007
Pages135-138
Number of pages4
StatePublished - 2007
Event8th International Conference on Music Information Retrieval, ISMIR 2007 - Vienna, Austria
Duration: Sep 23 2007Sep 27 2007

Other

Other8th International Conference on Music Information Retrieval, ISMIR 2007
CountryAustria
CityVienna
Period9/23/079/27/07

Fingerprint

Transducers
Support vector machines
Classifiers
Song
Music
Support Vector Machine
Mixture Model
Classifier

ASJC Scopus subject areas

  • Music
  • Information Systems

Cite this

Mohri, M., Moreno, P., & Weinstein, E. (2007). Robust music identification, detection, and analysis. In Proceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007 (pp. 135-138)

Robust music identification, detection, and analysis. / Mohri, Mehryar; Moreno, Pedro; Weinstein, Eugene.

Proceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007. 2007. p. 135-138.

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

Mohri, M, Moreno, P & Weinstein, E 2007, Robust music identification, detection, and analysis. in Proceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007. pp. 135-138, 8th International Conference on Music Information Retrieval, ISMIR 2007, Vienna, Austria, 9/23/07.
Mohri M, Moreno P, Weinstein E. Robust music identification, detection, and analysis. In Proceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007. 2007. p. 135-138
Mohri, Mehryar ; Moreno, Pedro ; Weinstein, Eugene. / Robust music identification, detection, and analysis. Proceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007. 2007. pp. 135-138
@inproceedings{232bb268ebfc4d339aaeac0f5dc5463a,
title = "Robust music identification, detection, and analysis",
abstract = "In previous work, we presented a new approach to music identification based on finite-state transducers and Gaussian mixture models. Here, we expand this work and study the performance of our system in the presence of noise and distortions. We also evaluate a song detection method based on a universal background model in combination with a support vector machine classifier and provide some insight into why our transducer representation allows for accurate identification even when only a short song snippet is available.",
author = "Mehryar Mohri and Pedro Moreno and Eugene Weinstein",
year = "2007",
language = "English (US)",
isbn = "9783854032182",
pages = "135--138",
booktitle = "Proceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007",

}

TY - GEN

T1 - Robust music identification, detection, and analysis

AU - Mohri, Mehryar

AU - Moreno, Pedro

AU - Weinstein, Eugene

PY - 2007

Y1 - 2007

N2 - In previous work, we presented a new approach to music identification based on finite-state transducers and Gaussian mixture models. Here, we expand this work and study the performance of our system in the presence of noise and distortions. We also evaluate a song detection method based on a universal background model in combination with a support vector machine classifier and provide some insight into why our transducer representation allows for accurate identification even when only a short song snippet is available.

AB - In previous work, we presented a new approach to music identification based on finite-state transducers and Gaussian mixture models. Here, we expand this work and study the performance of our system in the presence of noise and distortions. We also evaluate a song detection method based on a universal background model in combination with a support vector machine classifier and provide some insight into why our transducer representation allows for accurate identification even when only a short song snippet is available.

UR - http://www.scopus.com/inward/record.url?scp=67651107652&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67651107652&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783854032182

SP - 135

EP - 138

BT - Proceedings of the 8th International Conference on Music Information Retrieval, ISMIR 2007

ER -