İstatistiksel Sinir Aǧlari İle Konuşma/Müzik Siniflandirma

Bülent Bolat; Ünal Küçük

İstatistiksel Sinir Aǧlari İle Konuşma/Müzik Siniflandirma

Translated title of the contribution: Speeh/music classification by using statistical neural networks

Bülent Bolat^*, Ünal Küçük

^*Corresponding author for this work

Yildiz Technical University

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

4 Citations (Scopus)

Abstract

This paper represents a framework for speech/music classification by using statistical neural networks. Zero Crossing Rate, Root Mean Square Power and Spectral Centroid were used as features. A dataset include 150 audio instances were labeled manually and 105 of them were used to train different networks which are PNN, GRNN and RBF. The remaining of the dataset were used as test item. Training and test performances of these three network types were discussed.

Translated title of the contribution	Speeh/music classification by using statistical neural networks
Original language	Turkish
Title of host publication	Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004
Editors	B. Gunsel
Pages	227-229
Number of pages	3
Publication status	Published - 2004
Externally published	Yes
Event	Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004 - Kusadasi, Turkey Duration: 28 Apr 2004 → 30 Apr 2004

Publication series

Name	Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004

Conference

Conference	Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004
Country/Territory	Turkey
City	Kusadasi
Period	28/04/04 → 30/04/04

Cite this

@inproceedings{d8b4ee399071456f817a67675b83197c,

title = "İstatistiksel Sinir Aǧlari İle Konu{\c s}ma/M{\"u}zik Siniflandirma",

abstract = "This paper represents a framework for speech/music classification by using statistical neural networks. Zero Crossing Rate, Root Mean Square Power and Spectral Centroid were used as features. A dataset include 150 audio instances were labeled manually and 105 of them were used to train different networks which are PNN, GRNN and RBF. The remaining of the dataset were used as test item. Training and test performances of these three network types were discussed.",

author = "B{\"u}lent Bolat and {\"U}nal K{\"u}{\c c}{\"u}k",

year = "2004",

language = "T{\"u}rk{\c c}e",

isbn = "0780383184",

series = "Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004",

pages = "227--229",

editor = "B. Gunsel",

booktitle = "Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004",

note = "Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004 ; Conference date: 28-04-2004 Through 30-04-2004",

}

Bolat, B & Küçük, Ü 2004, İstatistiksel Sinir Aǧlari İle Konuşma/Müzik Siniflandirma. in B Gunsel (ed.), Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004. Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004, pp. 227-229, Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004, Kusadasi, Turkey, 28/04/04.

İstatistiksel Sinir Aǧlari İle Konuşma/Müzik Siniflandirma. / Bolat, Bülent; Küçük, Ünal.
Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004. ed. / B. Gunsel. 2004. p. 227-229 (Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004).

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

TY - GEN

T1 - İstatistiksel Sinir Aǧlari İle Konuşma/Müzik Siniflandirma

AU - Bolat, Bülent

AU - Küçük, Ünal

PY - 2004

Y1 - 2004

N2 - This paper represents a framework for speech/music classification by using statistical neural networks. Zero Crossing Rate, Root Mean Square Power and Spectral Centroid were used as features. A dataset include 150 audio instances were labeled manually and 105 of them were used to train different networks which are PNN, GRNN and RBF. The remaining of the dataset were used as test item. Training and test performances of these three network types were discussed.

AB - This paper represents a framework for speech/music classification by using statistical neural networks. Zero Crossing Rate, Root Mean Square Power and Spectral Centroid were used as features. A dataset include 150 audio instances were labeled manually and 105 of them were used to train different networks which are PNN, GRNN and RBF. The remaining of the dataset were used as test item. Training and test performances of these three network types were discussed.

UR - http://www.scopus.com/inward/record.url?scp=18844395654&partnerID=8YFLogxK

M3 - Konferans katkısı

AN - SCOPUS:18844395654

SN - 0780383184

T3 - Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004

SP - 227

EP - 229

BT - Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004

A2 - Gunsel, B.

T2 - Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004

Y2 - 28 April 2004 through 30 April 2004

ER -

İstatistiksel Sinir Aǧlari İle Konuşma/Müzik Siniflandirma

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this