Template-based and HMM-based Approaches for Isolated Spanish Digit Recognition

Fuente: Redalyc
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Lucas D. Terissi
Format: Artículo científico
Sprache:en
Veröffentlicht: Asociación Española para la Inteligencia Artificial 2005
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1876437326480015360
author Lucas D. Terissi
author_facet Lucas D. Terissi
contents Template-based and HMM-based Approaches for Isolated Spanish Digit Recognition Lucas D. Terissi Juan C. Gómez Ingeniería Cepstral features Hidden Markov Models Dynamic Time Warping IsolatedWord Recognition Isolated word recognition is usually performed by two diferent approaches, viz., Conventional Templatebasedand Hidden Markov Models (HMM)-based. A comparison between these approaches for isolated digitrecognition (in Spanish language) is performed in this paper. For the template-based approach, severalDynamic Time Warping algorithms are proposed and implemented in a Matlab environment, while for theHMM-based approach, the algorithms available in a freeware toolbox has been employed. The algorithmsare tested on a database collected during a Course on Digital Processing of Speech Signals at the Universityof Rosario. The test results show that a better performance can be obtained with a HMM-based digitrecognizer. 2005 artículo científico 1137-3601 https://www.redalyc.org/articulo.oa?id=92526910 en http://www.redalyc.org/revista.oa?id=925 Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial application/pdf Asociación Española para la Inteligencia Artificial Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial (España) Num.26 Vol.9
format Artículo científico
id redalyc_92526910
institution Redalyc
language en
publishDate 2005
publisher Asociación Española para la Inteligencia Artificial
spellingShingle Template-based and HMM-based Approaches for Isolated Spanish Digit Recognition
Lucas D. Terissi
Ingeniería
Cepstral features
Hidden Markov Models
Dynamic Time Warping
IsolatedWord Recognition
Template-based and HMM-based Approaches for Isolated Spanish Digit Recognition Lucas D. Terissi Juan C. Gómez Ingeniería Cepstral features Hidden Markov Models Dynamic Time Warping IsolatedWord Recognition Isolated word recognition is usually performed by two diferent approaches, viz., Conventional Templatebasedand Hidden Markov Models (HMM)-based. A comparison between these approaches for isolated digitrecognition (in Spanish language) is performed in this paper. For the template-based approach, severalDynamic Time Warping algorithms are proposed and implemented in a Matlab environment, while for theHMM-based approach, the algorithms available in a freeware toolbox has been employed. The algorithmsare tested on a database collected during a Course on Digital Processing of Speech Signals at the Universityof Rosario. The test results show that a better performance can be obtained with a HMM-based digitrecognizer. 2005 artículo científico 1137-3601 https://www.redalyc.org/articulo.oa?id=92526910 en http://www.redalyc.org/revista.oa?id=925 Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial application/pdf Asociación Española para la Inteligencia Artificial Inteligencia Artificial. Revista Iberoamericana de Inteligencia Artificial (España) Num.26 Vol.9
title Template-based and HMM-based Approaches for Isolated Spanish Digit Recognition
topic Ingeniería
Cepstral features
Hidden Markov Models
Dynamic Time Warping
IsolatedWord Recognition
url https://www.redalyc.org/articulo.oa?id=92526910