Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

Arabic Speech Recognition by MFCC and Bionic Wavelet Transform Using a Multi Layer Perceptron for Voice Control


Affiliations
1 Unit Research of Signal Processing, Tunis, Tunisia
     

   Subscribe/Renew Journal


In this paper, we have proposed a new technique of Arabic speech recognition with mono-locutor and a reduced vocabulary. This technique consists at first step in using our proper speech database containing Arabic speech words which are recorded by a mono-locutor for a voice command. The second step consists in features extraction from those recorded words. The third step consists in classifying those extracted features. The features extraction is performed by computing at first, the Mel Frequency Cepstral Coefficients (MFCCs) from each recorded word, then the Bionic Wavelet Transform (BWT) was applied to the vector obtained from the concatenation of the obtained MFCCs. The obtained bionic wavelet coefficients were then concatenated to construct one input of a Multi-Layer Perceptual (MLP) used for features classification. In the MLP learning and test phases, we have used eleven Arabic words each if them was repeated twenty five times by the same locutor. A simulation program used to test the performance of the proposed technique showed a classification rate equals to 99.39%.


Keywords

Arabic Speech Recognition, Bionic Wavelet Transforms (BWT), Feature Extraction, Mel-Frequency Cepstral Coefficients (MFCC), Multi-Layer Perceptron (MLP).
User
Subscription Login to verify subscription
Notifications
Font Size

Abstract Views: 219

PDF Views: 1




  • Arabic Speech Recognition by MFCC and Bionic Wavelet Transform Using a Multi Layer Perceptron for Voice Control

Abstract Views: 219  |  PDF Views: 1

Authors

M. Ben Nasr
Unit Research of Signal Processing, Tunis, Tunisia
M. Talbi
Unit Research of Signal Processing, Tunis, Tunisia
A. Cherif
Unit Research of Signal Processing, Tunis, Tunisia

Abstract


In this paper, we have proposed a new technique of Arabic speech recognition with mono-locutor and a reduced vocabulary. This technique consists at first step in using our proper speech database containing Arabic speech words which are recorded by a mono-locutor for a voice command. The second step consists in features extraction from those recorded words. The third step consists in classifying those extracted features. The features extraction is performed by computing at first, the Mel Frequency Cepstral Coefficients (MFCCs) from each recorded word, then the Bionic Wavelet Transform (BWT) was applied to the vector obtained from the concatenation of the obtained MFCCs. The obtained bionic wavelet coefficients were then concatenated to construct one input of a Multi-Layer Perceptual (MLP) used for features classification. In the MLP learning and test phases, we have used eleven Arabic words each if them was repeated twenty five times by the same locutor. A simulation program used to test the performance of the proposed technique showed a classification rate equals to 99.39%.


Keywords


Arabic Speech Recognition, Bionic Wavelet Transforms (BWT), Feature Extraction, Mel-Frequency Cepstral Coefficients (MFCC), Multi-Layer Perceptron (MLP).