A Study of Perceptron Mapping Capability to Design Speech Event Detectors

Risultato della ricerca: Other

1 Citazione (Scopus)

Abstract

Event detection is a fundamental yet critical component in automatic speech recognition (ASR) systems that attempt to extract knowledge-based features at the front-end level. In this context, it is common practice to design the detectors inside well-known frameworks based on artificial neural network (ANN) or support vector machine (SVM). In the case of ANN, speech scientists often design their detector architecture relying on conventional feed-forward multi-layer perceptron (MLP) with sigmoidal activation function. The aim of this paper is to introduce other ANN architectures inside the context of detection-based ASR. In particular, a bank of feed-forward MLPs using sinusoidal activation functions is set up to address the event detection problem. Experimental results demonstrate the effectiveness of this ANN design for speech attribute detectors.
Lingua originaleEnglish
Pagine805-808
Numero di pagine4
Stato di pubblicazionePublished - 2006

Fingerprint

Detectors
Neural networks
Speech recognition
Chemical activation
Multilayer neural networks
Network architecture
Support vector machines

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Cita questo

@conference{dbbfbf7fc2ba4a3aa4a82e4dfeb2638a,
title = "A Study of Perceptron Mapping Capability to Design Speech Event Detectors",
abstract = "Event detection is a fundamental yet critical component in automatic speech recognition (ASR) systems that attempt to extract knowledge-based features at the front-end level. In this context, it is common practice to design the detectors inside well-known frameworks based on artificial neural network (ANN) or support vector machine (SVM). In the case of ANN, speech scientists often design their detector architecture relying on conventional feed-forward multi-layer perceptron (MLP) with sigmoidal activation function. The aim of this paper is to introduce other ANN architectures inside the context of detection-based ASR. In particular, a bank of feed-forward MLPs using sinusoidal activation functions is set up to address the event detection problem. Experimental results demonstrate the effectiveness of this ANN design for speech attribute detectors.",
keywords = "Speech, recognition, speech segmentation",
author = "Filippo Sorbello and Giorgio Vassallo and Siniscalchi, {Sabato Marco} and Antonio Gentile and Siniscalchi, {Sabato M.} and Clements, {Mark A.}",
year = "2006",
language = "English",
pages = "805--808",

}

TY - CONF

T1 - A Study of Perceptron Mapping Capability to Design Speech Event Detectors

AU - Sorbello, Filippo

AU - Vassallo, Giorgio

AU - Siniscalchi, Sabato Marco

AU - Gentile, Antonio

AU - Siniscalchi, Sabato M.

AU - Clements, Mark A.

PY - 2006

Y1 - 2006

N2 - Event detection is a fundamental yet critical component in automatic speech recognition (ASR) systems that attempt to extract knowledge-based features at the front-end level. In this context, it is common practice to design the detectors inside well-known frameworks based on artificial neural network (ANN) or support vector machine (SVM). In the case of ANN, speech scientists often design their detector architecture relying on conventional feed-forward multi-layer perceptron (MLP) with sigmoidal activation function. The aim of this paper is to introduce other ANN architectures inside the context of detection-based ASR. In particular, a bank of feed-forward MLPs using sinusoidal activation functions is set up to address the event detection problem. Experimental results demonstrate the effectiveness of this ANN design for speech attribute detectors.

AB - Event detection is a fundamental yet critical component in automatic speech recognition (ASR) systems that attempt to extract knowledge-based features at the front-end level. In this context, it is common practice to design the detectors inside well-known frameworks based on artificial neural network (ANN) or support vector machine (SVM). In the case of ANN, speech scientists often design their detector architecture relying on conventional feed-forward multi-layer perceptron (MLP) with sigmoidal activation function. The aim of this paper is to introduce other ANN architectures inside the context of detection-based ASR. In particular, a bank of feed-forward MLPs using sinusoidal activation functions is set up to address the event detection problem. Experimental results demonstrate the effectiveness of this ANN design for speech attribute detectors.

KW - Speech

KW - recognition

KW - speech segmentation

UR - http://hdl.handle.net/10447/34281

M3 - Other

SP - 805

EP - 808

ER -