Warping Indexes with Envelope Transforms for Query by Humming

Yunyue Zhu, Dennis Shasha

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

Abstract

A Query by Humming system allows the user to find a song by humming part of the tune. No musical training is needed. Previous query by humming systems have not provided satisfactory results for various reasons. Some systems have low retrieval precision because they rely on melodic contour information from the hum tune, which in turn relies on the error-prone note segmentation process. Some systems yield better precision when matching the melody directly from audio, but they are slow because of their extensive use of Dynamic Time Warping (DTW). Our approach improves both the retrieval precision and speed compared to previous approaches. We treat music as a time series and exploit and improve well-developed techniques from time series databases to index the music for fast similarity queries. We improve on existing DTW indexes technique by introducing the concept of envelope transforms, which gives a general guideline for extending existing dimensionality reduction methods to DTW indexes. The net result is high scalability. We confirm our claims through extensive experiments.

Original languageEnglish (US)
Title of host publicationProceedings of the ACM SIGMOD International Conference on Management of Data
EditorsA.Y. Halevy, Z.G. Ives, A.H. Doan
Pages181-192
Number of pages12
StatePublished - 2003
Event2003 ACM SIGMOD International Conference on Management of Data - San Diego, CA, United States
Duration: Jun 9 2003Jun 12 2003

Other

Other2003 ACM SIGMOD International Conference on Management of Data
CountryUnited States
CitySan Diego, CA
Period6/9/036/12/03

Fingerprint

Mathematical transformations
Time series
Scalability
Experiments

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Zhu, Y., & Shasha, D. (2003). Warping Indexes with Envelope Transforms for Query by Humming. In A. Y. Halevy, Z. G. Ives, & A. H. Doan (Eds.), Proceedings of the ACM SIGMOD International Conference on Management of Data (pp. 181-192)

Warping Indexes with Envelope Transforms for Query by Humming. / Zhu, Yunyue; Shasha, Dennis.

Proceedings of the ACM SIGMOD International Conference on Management of Data. ed. / A.Y. Halevy; Z.G. Ives; A.H. Doan. 2003. p. 181-192.

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

Zhu, Y & Shasha, D 2003, Warping Indexes with Envelope Transforms for Query by Humming. in AY Halevy, ZG Ives & AH Doan (eds), Proceedings of the ACM SIGMOD International Conference on Management of Data. pp. 181-192, 2003 ACM SIGMOD International Conference on Management of Data, San Diego, CA, United States, 6/9/03.
Zhu Y, Shasha D. Warping Indexes with Envelope Transforms for Query by Humming. In Halevy AY, Ives ZG, Doan AH, editors, Proceedings of the ACM SIGMOD International Conference on Management of Data. 2003. p. 181-192
Zhu, Yunyue ; Shasha, Dennis. / Warping Indexes with Envelope Transforms for Query by Humming. Proceedings of the ACM SIGMOD International Conference on Management of Data. editor / A.Y. Halevy ; Z.G. Ives ; A.H. Doan. 2003. pp. 181-192
@inproceedings{02230fb41ab444fc852b87383db688a1,
title = "Warping Indexes with Envelope Transforms for Query by Humming",
abstract = "A Query by Humming system allows the user to find a song by humming part of the tune. No musical training is needed. Previous query by humming systems have not provided satisfactory results for various reasons. Some systems have low retrieval precision because they rely on melodic contour information from the hum tune, which in turn relies on the error-prone note segmentation process. Some systems yield better precision when matching the melody directly from audio, but they are slow because of their extensive use of Dynamic Time Warping (DTW). Our approach improves both the retrieval precision and speed compared to previous approaches. We treat music as a time series and exploit and improve well-developed techniques from time series databases to index the music for fast similarity queries. We improve on existing DTW indexes technique by introducing the concept of envelope transforms, which gives a general guideline for extending existing dimensionality reduction methods to DTW indexes. The net result is high scalability. We confirm our claims through extensive experiments.",
author = "Yunyue Zhu and Dennis Shasha",
year = "2003",
language = "English (US)",
pages = "181--192",
editor = "A.Y. Halevy and Z.G. Ives and A.H. Doan",
booktitle = "Proceedings of the ACM SIGMOD International Conference on Management of Data",

}

TY - GEN

T1 - Warping Indexes with Envelope Transforms for Query by Humming

AU - Zhu, Yunyue

AU - Shasha, Dennis

PY - 2003

Y1 - 2003

N2 - A Query by Humming system allows the user to find a song by humming part of the tune. No musical training is needed. Previous query by humming systems have not provided satisfactory results for various reasons. Some systems have low retrieval precision because they rely on melodic contour information from the hum tune, which in turn relies on the error-prone note segmentation process. Some systems yield better precision when matching the melody directly from audio, but they are slow because of their extensive use of Dynamic Time Warping (DTW). Our approach improves both the retrieval precision and speed compared to previous approaches. We treat music as a time series and exploit and improve well-developed techniques from time series databases to index the music for fast similarity queries. We improve on existing DTW indexes technique by introducing the concept of envelope transforms, which gives a general guideline for extending existing dimensionality reduction methods to DTW indexes. The net result is high scalability. We confirm our claims through extensive experiments.

AB - A Query by Humming system allows the user to find a song by humming part of the tune. No musical training is needed. Previous query by humming systems have not provided satisfactory results for various reasons. Some systems have low retrieval precision because they rely on melodic contour information from the hum tune, which in turn relies on the error-prone note segmentation process. Some systems yield better precision when matching the melody directly from audio, but they are slow because of their extensive use of Dynamic Time Warping (DTW). Our approach improves both the retrieval precision and speed compared to previous approaches. We treat music as a time series and exploit and improve well-developed techniques from time series databases to index the music for fast similarity queries. We improve on existing DTW indexes technique by introducing the concept of envelope transforms, which gives a general guideline for extending existing dimensionality reduction methods to DTW indexes. The net result is high scalability. We confirm our claims through extensive experiments.

UR - http://www.scopus.com/inward/record.url?scp=1142279463&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1142279463&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:1142279463

SP - 181

EP - 192

BT - Proceedings of the ACM SIGMOD International Conference on Management of Data

A2 - Halevy, A.Y.

A2 - Ives, Z.G.

A2 - Doan, A.H.

ER -