Fast query by example of environmental sounds via robust and efficient cluster-based indexing

Jiachen Xue, Gordon Wichern, Harvey Thornburg, Andreas Spanias

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

14 Citations (Scopus)

Abstract

There has been much recent progress in the technical infrastructure necessary to continuously characterize and archive all sounds, or more precisely auditory streams, that occur within a given space or human life. Efficient and intuitive access, however, remains a considerable challenge. In specifically musical domains, i.e., melody retrieval, query-by-example (QBE) has found considerable success in accessing music that matches a specific query. We propose an extension of the QBE paradigm to the broad class of natural and environmental sounds, which occur frequently in continuous recordings. We explore several cluster-based indexing approaches, namely non-negative matrix factorization (NMF) and spectral clustering to efficiently organize and quickly retrieve archived audio using the QBE paradigm. Experiments on a test database compare the performance of the different clustering algorithms in terms of recall, precision, and computational complexity. Initial results indicate significant improvements over both exhaustive search schemes and traditional K-means clustering, and excellent overall performance in the example-based retrieval of environmental sounds.

Original languageEnglish (US)
Title of host publicationICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Pages5-8
Number of pages4
DOIs
StatePublished - 2008
Event2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP - Las Vegas, NV, United States
Duration: Mar 31 2008Apr 4 2008

Other

Other2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP
CountryUnited States
CityLas Vegas, NV
Period3/31/084/4/08

Fingerprint

Acoustic waves
retrieval
acoustics
music
Factorization
factorization
Clustering algorithms
Computational complexity
recording
matrices
Experiments

Keywords

  • Acoustic signal analysis
  • Clustering methods
  • Database query processing
  • Hidden Markov models

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Signal Processing
  • Acoustics and Ultrasonics

Cite this

Xue, J., Wichern, G., Thornburg, H., & Spanias, A. (2008). Fast query by example of environmental sounds via robust and efficient cluster-based indexing. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings (pp. 5-8). [4517532] https://doi.org/10.1109/ICASSP.2008.4517532

Fast query by example of environmental sounds via robust and efficient cluster-based indexing. / Xue, Jiachen; Wichern, Gordon; Thornburg, Harvey; Spanias, Andreas.

ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2008. p. 5-8 4517532.

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

Xue, J, Wichern, G, Thornburg, H & Spanias, A 2008, Fast query by example of environmental sounds via robust and efficient cluster-based indexing. in ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings., 4517532, pp. 5-8, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP, Las Vegas, NV, United States, 3/31/08. https://doi.org/10.1109/ICASSP.2008.4517532
Xue J, Wichern G, Thornburg H, Spanias A. Fast query by example of environmental sounds via robust and efficient cluster-based indexing. In ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2008. p. 5-8. 4517532 https://doi.org/10.1109/ICASSP.2008.4517532
Xue, Jiachen ; Wichern, Gordon ; Thornburg, Harvey ; Spanias, Andreas. / Fast query by example of environmental sounds via robust and efficient cluster-based indexing. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. 2008. pp. 5-8
@inproceedings{aa6017d550064a619485764546d78f2f,
title = "Fast query by example of environmental sounds via robust and efficient cluster-based indexing",
abstract = "There has been much recent progress in the technical infrastructure necessary to continuously characterize and archive all sounds, or more precisely auditory streams, that occur within a given space or human life. Efficient and intuitive access, however, remains a considerable challenge. In specifically musical domains, i.e., melody retrieval, query-by-example (QBE) has found considerable success in accessing music that matches a specific query. We propose an extension of the QBE paradigm to the broad class of natural and environmental sounds, which occur frequently in continuous recordings. We explore several cluster-based indexing approaches, namely non-negative matrix factorization (NMF) and spectral clustering to efficiently organize and quickly retrieve archived audio using the QBE paradigm. Experiments on a test database compare the performance of the different clustering algorithms in terms of recall, precision, and computational complexity. Initial results indicate significant improvements over both exhaustive search schemes and traditional K-means clustering, and excellent overall performance in the example-based retrieval of environmental sounds.",
keywords = "Acoustic signal analysis, Clustering methods, Database query processing, Hidden Markov models",
author = "Jiachen Xue and Gordon Wichern and Harvey Thornburg and Andreas Spanias",
year = "2008",
doi = "10.1109/ICASSP.2008.4517532",
language = "English (US)",
isbn = "1424414849",
pages = "5--8",
booktitle = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

}

TY - GEN

T1 - Fast query by example of environmental sounds via robust and efficient cluster-based indexing

AU - Xue, Jiachen

AU - Wichern, Gordon

AU - Thornburg, Harvey

AU - Spanias, Andreas

PY - 2008

Y1 - 2008

N2 - There has been much recent progress in the technical infrastructure necessary to continuously characterize and archive all sounds, or more precisely auditory streams, that occur within a given space or human life. Efficient and intuitive access, however, remains a considerable challenge. In specifically musical domains, i.e., melody retrieval, query-by-example (QBE) has found considerable success in accessing music that matches a specific query. We propose an extension of the QBE paradigm to the broad class of natural and environmental sounds, which occur frequently in continuous recordings. We explore several cluster-based indexing approaches, namely non-negative matrix factorization (NMF) and spectral clustering to efficiently organize and quickly retrieve archived audio using the QBE paradigm. Experiments on a test database compare the performance of the different clustering algorithms in terms of recall, precision, and computational complexity. Initial results indicate significant improvements over both exhaustive search schemes and traditional K-means clustering, and excellent overall performance in the example-based retrieval of environmental sounds.

AB - There has been much recent progress in the technical infrastructure necessary to continuously characterize and archive all sounds, or more precisely auditory streams, that occur within a given space or human life. Efficient and intuitive access, however, remains a considerable challenge. In specifically musical domains, i.e., melody retrieval, query-by-example (QBE) has found considerable success in accessing music that matches a specific query. We propose an extension of the QBE paradigm to the broad class of natural and environmental sounds, which occur frequently in continuous recordings. We explore several cluster-based indexing approaches, namely non-negative matrix factorization (NMF) and spectral clustering to efficiently organize and quickly retrieve archived audio using the QBE paradigm. Experiments on a test database compare the performance of the different clustering algorithms in terms of recall, precision, and computational complexity. Initial results indicate significant improvements over both exhaustive search schemes and traditional K-means clustering, and excellent overall performance in the example-based retrieval of environmental sounds.

KW - Acoustic signal analysis

KW - Clustering methods

KW - Database query processing

KW - Hidden Markov models

UR - http://www.scopus.com/inward/record.url?scp=51449089882&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=51449089882&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2008.4517532

DO - 10.1109/ICASSP.2008.4517532

M3 - Conference contribution

AN - SCOPUS:51449089882

SN - 1424414849

SN - 9781424414840

SP - 5

EP - 8

BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

ER -