Robust Arabic speech recognition in noisy environments using prosodic features and formant

Abstract

This paper investigates the contribution of formants and prosodic features such as pitch and energy in Arabic speech recognition under real-life conditions. Our speech recognition system based on Hidden Markov Models (HMMs) is implemented using the HTK Toolkit. The frontend of the system combines features based on conventional Mel-Frequency Cepstral Coefficient (MFFC), prosodic information and formants. The experiments are performed on the ARADIGIT corpus which is a database of Arabic spoken words. The obtained results show that the resulting multivariate feature vectors, in noisy environment, lead to a significant improvement, up to 27%, in word accuracy relative the word accuracy obtained from the state-of-the-art MFCCbased system.

Arabic digits speech recognition and speaker identification in noisy environment using a hybrid model of VQ and GMM

“…The Arabic language is the fourth largest language spoken by nearly 1.6 billion Muslims native speakers, this language spoken by the majority of the people in the Middle East and North Africa; note that Arabic has many different dialects. This is some little work in Arabic speaker and speech recognition [35][36][37][38]. We presented in Table 1 a Literature review for speech recognition research using MFCC, GMM and VQ techniques regarding Arabic or other languages.…”