Pitch Based Sound Classification

Andreas Brinch Nielsen; Lars Kai Hansen; U Kjems

doi:10.1109/ICASSP.2006.1660772

Pitch Based Sound Classification

Andreas Brinch Nielsen, Lars Kai Hansen, U Kjems

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

1110 Downloads (Orbit)

Abstract

A sound classification model is presented that can classify signals into music, noise and speech. The model extracts the pitch of the signal using the harmonic product spectrum. Based on the pitch estimate and a pitch error measure, features are created and used in a probabilistic model with soft-max output function. Both linear and quadratic inputs are used. The model is trained on 2 hours of sound and tested on publicly available data. A test classification error below 0.05 with 1 s classification windows is achieved. Further more it is shown that linear input performs as well as a quadratic, and that even though classification gets marginally better, not much is achieved by increasing the window size beyond 1 s.

Original language	English
Title of host publication	2006 IEEE International Conference on Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings.
Volume	3
Publisher	IEEE
Publication date	2006
ISBN (Print)	1-4244-0469-X
DOIs	https://doi.org/10.1109/ICASSP.2006.1660772
Publication status	Published - 2006
Event	2006 IEEE International Conference on Acoustics, Speech and Signal Processing - Toulouse, France Duration: 14 May 2006 → 19 May 2006 Conference number: 31

Conference

Conference	2006 IEEE International Conference on Acoustics, Speech and Signal Processing
Number	31
Country/Territory	France
City	Toulouse
Period	14/05/2006 → 19/05/2006

Bibliographical note

Copyright: 2006 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

Access to Document

10.1109/ICASSP.2006.1660772

Nielsen2Final published version, 2.05 MB

OpenUrl availability

Full text

Cite this

@inproceedings{f14fe651fc8a45ec8b37d84c48eb970e,

title = "Pitch Based Sound Classification",

abstract = "A sound classification model is presented that can classify signals into music, noise and speech. The model extracts the pitch of the signal using the harmonic product spectrum. Based on the pitch estimate and a pitch error measure, features are created and used in a probabilistic model with soft-max output function. Both linear and quadratic inputs are used. The model is trained on 2 hours of sound and tested on publicly available data. A test classification error below 0.05 with 1 s classification windows is achieved. Further more it is shown that linear input performs as well as a quadratic, and that even though classification gets marginally better, not much is achieved by increasing the window size beyond 1 s.",

author = "Nielsen, {Andreas Brinch} and Hansen, {Lars Kai} and U Kjems",

note = "Copyright: 2006 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE; 2006 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2006 ; Conference date: 14-05-2006 Through 19-05-2006",

year = "2006",

doi = "10.1109/ICASSP.2006.1660772",

language = "English",

isbn = "1-4244-0469-X",

volume = "3",

booktitle = "2006 IEEE International Conference on Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings.",

publisher = "IEEE",

address = "United States",

}

TY - GEN

T1 - Pitch Based Sound Classification

AU - Nielsen, Andreas Brinch

AU - Hansen, Lars Kai

AU - Kjems, U

N1 - Conference code: 31

PY - 2006

Y1 - 2006

N2 - A sound classification model is presented that can classify signals into music, noise and speech. The model extracts the pitch of the signal using the harmonic product spectrum. Based on the pitch estimate and a pitch error measure, features are created and used in a probabilistic model with soft-max output function. Both linear and quadratic inputs are used. The model is trained on 2 hours of sound and tested on publicly available data. A test classification error below 0.05 with 1 s classification windows is achieved. Further more it is shown that linear input performs as well as a quadratic, and that even though classification gets marginally better, not much is achieved by increasing the window size beyond 1 s.

AB - A sound classification model is presented that can classify signals into music, noise and speech. The model extracts the pitch of the signal using the harmonic product spectrum. Based on the pitch estimate and a pitch error measure, features are created and used in a probabilistic model with soft-max output function. Both linear and quadratic inputs are used. The model is trained on 2 hours of sound and tested on publicly available data. A test classification error below 0.05 with 1 s classification windows is achieved. Further more it is shown that linear input performs as well as a quadratic, and that even though classification gets marginally better, not much is achieved by increasing the window size beyond 1 s.

U2 - 10.1109/ICASSP.2006.1660772

DO - 10.1109/ICASSP.2006.1660772

M3 - Article in proceedings

SN - 1-4244-0469-X

VL - 3

BT - 2006 IEEE International Conference on Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings.

PB - IEEE

T2 - 2006 IEEE International Conference on Acoustics, Speech and Signal Processing

Y2 - 14 May 2006 through 19 May 2006

ER -

Pitch Based Sound Classification

Abstract

Conference

Bibliographical note

Access to Document

OpenUrl availability

Fingerprint

Cite this