Comparative study and improvement of acoustic vectors extractors: Multiple streams applied to the recognition of Arabic numerals

Author

Lamkadam, Abdelmajid ; Karim, Mohamed

Author_Institution

Fac. of Sci. Dhar El Mehraz, Dept. of Math. & Comput. Sci., Univ. Sidi Mohamed Ben Abdellah, Fez, Morocco

fYear

2015

fDate

25-26 March 2015

Firstpage

1

Lastpage

9

Abstract

This paper exposes at the beginning most of the problems encountered in the automatic speech recognition, namely the complexity of speech signal, and the diversity of process of signal processing. For this, a review of general concepts in signal processing is inspected; after it has been specifically addressed the extraction phase, in order to conduct a thorough review of existing extraction methods in this area. The next task is reserved to a list of existing extractor´s combinations, and finally an extractor´s recombination is developed, with the aim of reaching a full treatment with a considerable amount of extracted vectors acoustic, in order to develop the recognition rate of Arabic numerals.

Keywords

acoustic signal processing; speech recognition; Arabic numeral recognition rate; acoustic vector extractor improvement; automatic speech recognition; extractor recombination; signal processing diversity; speech signal complexity; Mel frequency cepstral coefficient; Robustness; Speech; Speech processing; Speech recognition; Automatic Speech Recognition (ASR); Combination; DFE; DTW; LDA; LPC; MFCC; PLP; RASTA; acoustic vectors; extraction methods; speech signal processing;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Systems and Computer Vision (ISCV), 2015

Conference_Location

Fez

Print_ISBN

978-1-4799-7510-5

Type

conf

DOI

10.1109/ISACV.2015.7106179

Filename

7106179