On the Reduction of False Positives in Singing Voice Detection

B. Lehner; Gerhard Widmer; Reinhard Sonnleitner

On the Reduction of False Positives in Singing Voice Detection

B. Lehner, Gerhard Widmer, Reinhard Sonnleitner

Institute of Computational Perception

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

Abstract

Motivated by the observation that one of the biggest problems in automatic singing voice detection is the confusion of vocals with other pitch-continuous and pitch-varying instruments, we propose a set of three new audio features designed to reduce the amount of false vocal detections. This is borne out in comparative experiments with three different musical corpora. The resulting singing voice detector appears to be at least on par with more complex state-of-the-art methods. New features and classifier are very light-weight and in principle suitable for on-line use.

Original language	English
Title of host publication	39th International Conference on Acoustics, Speech, and Signal Processing (ICASSP) , Florence, Italy.
Number of pages	5
Publication status	Published - 2014

Fields of science

202002 Audiovisual media
102 Computer Sciences
102001 Artificial intelligence
102003 Image processing
102015 Information systems

JKU Focus areas

Computation in Informatics and Mathematics
Engineering and Natural Sciences (in general)

Cite this

@inproceedings{313cc4a2d3364e03a920ce9642c0a31c,

title = "On the Reduction of False Positives in Singing Voice Detection",

abstract = "Motivated by the observation that one of the biggest problems in automatic singing voice detection is the confusion of vocals with other pitch-continuous and pitch-varying instruments, we propose a set of three new audio features designed to reduce the amount of false vocal detections. This is borne out in comparative experiments with three different musical corpora. The resulting singing voice detector appears to be at least on par with more complex state-of-the-art methods. New features and classifier are very light-weight and in principle suitable for on-line use.",

author = "B. Lehner and Gerhard Widmer and Reinhard Sonnleitner",

year = "2014",

language = "English",

booktitle = "39th International Conference on Acoustics, Speech, and Signal Processing (ICASSP) , Florence, Italy.",

}

TY - GEN

T1 - On the Reduction of False Positives in Singing Voice Detection

AU - Lehner, B.

AU - Widmer, Gerhard

AU - Sonnleitner, Reinhard

PY - 2014

Y1 - 2014

N2 - Motivated by the observation that one of the biggest problems in automatic singing voice detection is the confusion of vocals with other pitch-continuous and pitch-varying instruments, we propose a set of three new audio features designed to reduce the amount of false vocal detections. This is borne out in comparative experiments with three different musical corpora. The resulting singing voice detector appears to be at least on par with more complex state-of-the-art methods. New features and classifier are very light-weight and in principle suitable for on-line use.

AB - Motivated by the observation that one of the biggest problems in automatic singing voice detection is the confusion of vocals with other pitch-continuous and pitch-varying instruments, we propose a set of three new audio features designed to reduce the amount of false vocal detections. This is borne out in comparative experiments with three different musical corpora. The resulting singing voice detector appears to be at least on par with more complex state-of-the-art methods. New features and classifier are very light-weight and in principle suitable for on-line use.

M3 - Conference proceedings

BT - 39th International Conference on Acoustics, Speech, and Signal Processing (ICASSP) , Florence, Italy.

ER -