Speech recognition with a competitive probabilistic radial basis neural network

Nima Yousefian; Azarakhsh Jalalvand; Pooyan Ahmadi; Morteza Analoui

doi:10.1109/IS.2008.4670445

Speech recognition with a competitive probabilistic radial basis neural network

Nima Yousefian, Azarakhsh Jalalvand, Pooyan Ahmadi, Morteza Analoui

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

3 Scopus citations

Abstract

Automatic speech recognition (ASR) has been subject of active research in the last few decades. In this paper we study the applicability of a special model of radial basis probabilistic neural networks (RBPNN) as a classifier for speech recognition. This type of network is a combination Radial Basis Function (RBF) and Probabilistic Neural Network (PNN) that applies characteristics of both networks and finally uses a competitive function for computing final result. The proposed network has been tested on the voices single digit numbers dataset and produced lower recognition error rate in comparison with other common pattern classifiers. All of classifiers use Mel-scale Frequency Cepstrum Coefficients (MFCC) and a special type Perceptual Linear Predictive (PLP) as their features for classification. Results show that PLP features yield better recognition rate by considering noisy dataset.

Original language	English (US)
Title of host publication	2008 4th International IEEE Conference Intelligent Systems, IS 2008
Pages	719-723
Number of pages	5
DOIs	https://doi.org/10.1109/IS.2008.4670445
State	Published - 2008
Externally published	Yes
Event	2008 4th International IEEE Conference Intelligent Systems, IS 2008 - Varna, Bulgaria Duration: Sep 6 2008 → Sep 8 2008

Publication series

Name	2008 4th International IEEE Conference Intelligent Systems, IS 2008
Volume	1

Conference

Conference	2008 4th International IEEE Conference Intelligent Systems, IS 2008
Country/Territory	Bulgaria
City	Varna
Period	9/6/08 → 9/8/08

All Science Journal Classification (ASJC) codes

Artificial Intelligence
Information Systems
Electrical and Electronic Engineering

Keywords

Probabilistic networks
Radial basis function
Speech recognition

Access to Document

10.1109/IS.2008.4670445

Cite this

@inproceedings{23d8da5c7d0b4ecf9b97d5250da1962f,

title = "Speech recognition with a competitive probabilistic radial basis neural network",

abstract = "Automatic speech recognition (ASR) has been subject of active research in the last few decades. In this paper we study the applicability of a special model of radial basis probabilistic neural networks (RBPNN) as a classifier for speech recognition. This type of network is a combination Radial Basis Function (RBF) and Probabilistic Neural Network (PNN) that applies characteristics of both networks and finally uses a competitive function for computing final result. The proposed network has been tested on the voices single digit numbers dataset and produced lower recognition error rate in comparison with other common pattern classifiers. All of classifiers use Mel-scale Frequency Cepstrum Coefficients (MFCC) and a special type Perceptual Linear Predictive (PLP) as their features for classification. Results show that PLP features yield better recognition rate by considering noisy dataset.",

keywords = "Probabilistic networks, Radial basis function, Speech recognition",

author = "Nima Yousefian and Azarakhsh Jalalvand and Pooyan Ahmadi and Morteza Analoui",

year = "2008",

doi = "10.1109/IS.2008.4670445",

language = "English (US)",

isbn = "9781424417391",

series = "2008 4th International IEEE Conference Intelligent Systems, IS 2008",

pages = "719--723",

booktitle = "2008 4th International IEEE Conference Intelligent Systems, IS 2008",

note = "2008 4th International IEEE Conference Intelligent Systems, IS 2008 ; Conference date: 06-09-2008 Through 08-09-2008",

}

Yousefian, N, Jalalvand, A, Ahmadi, P & Analoui, M 2008, Speech recognition with a competitive probabilistic radial basis neural network. in 2008 4th International IEEE Conference Intelligent Systems, IS 2008., 4670445, 2008 4th International IEEE Conference Intelligent Systems, IS 2008, vol. 1, pp. 719-723, 2008 4th International IEEE Conference Intelligent Systems, IS 2008, Varna, Bulgaria, 9/6/08. https://doi.org/10.1109/IS.2008.4670445

Speech recognition with a competitive probabilistic radial basis neural network. / Yousefian, Nima; Jalalvand, Azarakhsh; Ahmadi, Pooyan et al.
2008 4th International IEEE Conference Intelligent Systems, IS 2008. 2008. p. 719-723 4670445 (2008 4th International IEEE Conference Intelligent Systems, IS 2008; Vol. 1).

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

TY - GEN

T1 - Speech recognition with a competitive probabilistic radial basis neural network

AU - Yousefian, Nima

AU - Jalalvand, Azarakhsh

AU - Ahmadi, Pooyan

AU - Analoui, Morteza

PY - 2008

Y1 - 2008

N2 - Automatic speech recognition (ASR) has been subject of active research in the last few decades. In this paper we study the applicability of a special model of radial basis probabilistic neural networks (RBPNN) as a classifier for speech recognition. This type of network is a combination Radial Basis Function (RBF) and Probabilistic Neural Network (PNN) that applies characteristics of both networks and finally uses a competitive function for computing final result. The proposed network has been tested on the voices single digit numbers dataset and produced lower recognition error rate in comparison with other common pattern classifiers. All of classifiers use Mel-scale Frequency Cepstrum Coefficients (MFCC) and a special type Perceptual Linear Predictive (PLP) as their features for classification. Results show that PLP features yield better recognition rate by considering noisy dataset.

AB - Automatic speech recognition (ASR) has been subject of active research in the last few decades. In this paper we study the applicability of a special model of radial basis probabilistic neural networks (RBPNN) as a classifier for speech recognition. This type of network is a combination Radial Basis Function (RBF) and Probabilistic Neural Network (PNN) that applies characteristics of both networks and finally uses a competitive function for computing final result. The proposed network has been tested on the voices single digit numbers dataset and produced lower recognition error rate in comparison with other common pattern classifiers. All of classifiers use Mel-scale Frequency Cepstrum Coefficients (MFCC) and a special type Perceptual Linear Predictive (PLP) as their features for classification. Results show that PLP features yield better recognition rate by considering noisy dataset.

KW - Probabilistic networks

KW - Radial basis function

KW - Speech recognition

UR - http://www.scopus.com/inward/record.url?scp=67249153104&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67249153104&partnerID=8YFLogxK

U2 - 10.1109/IS.2008.4670445

DO - 10.1109/IS.2008.4670445

M3 - Conference contribution

AN - SCOPUS:67249153104

SN - 9781424417391

T3 - 2008 4th International IEEE Conference Intelligent Systems, IS 2008

SP - 719

EP - 723

BT - 2008 4th International IEEE Conference Intelligent Systems, IS 2008

T2 - 2008 4th International IEEE Conference Intelligent Systems, IS 2008

Y2 - 6 September 2008 through 8 September 2008

ER -

Speech recognition with a competitive probabilistic radial basis neural network

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Keywords

Access to Document

Other files and links

Fingerprint

Cite this