Text-Dependent Multilingual Speaker Identification using Back Propagation Neural Network and PSO-GA Hybrid Model

In this work a multilingual speaker identification system is proposed. The feature extraction techniques employed in the system extract Mel frequency cepstral coefficient (MFCC), delta mel frequency cepstral coefficient (DMFCC) and format frequency. The feature selection is done using hybrid model of particle swarm optimizatiom (PSO) and Genetic Algorithm (GA). We have used Back Propagation (BPNN) artificial Neural Network classifiers. The speech database consists of 40 speakers(20 males+ 20 females) speech utterance. The speech utterance is recorded for a specific sentence in three different languages viz. “Now this time you go” (in English), “Adhuna Asmin Twam Gachh “(in sanskrit), “Ab Iss Baar Tum Jao” (in Hindi). Total word for this purpose is 14 including 4 for Sanskrit and 5 Hindi and English. The average identification rate 72.31% is achieved when the network is trained by BPNN and it shows 72.84% when BPNN is trained using hybrid PSO-GA model. Key words— Mel frequency cepstral coefficient (MFCC), delta mel frequency cepstral coefficient (DMFCC), format frequency, particle swarm optimizatiom (PSO), Genetic Algorithm (GA), Back Propagation Neural Network (BPNN))