Speaker Recognition using Particle Swarm Optimization (PSO)
Rita Yadav, Danvir Mandal · 2011
Human speech is the foundation of self-expression and communication with others. Speaker Recognition (SR) is the process of identifying a person based on his voice, as each person has different voice. In this paper we have implemented technique based on Particle Optimization (PSO) to improve the performance of speaker recognition. PSO is a search algorithm, in which each potential solution is seen as a particle with a certain velocity flying through the problem space. The Particle Swarms find optimal regions of the complex search space through the interaction of individuals in the population. PSO is attractive for optimization in that particle swarms will discover best optimized value as they fly within the subset space. In today world, the usage of digital systems is continuously growing high for daily dealings and communication among people and organizations. These organizations can be government, public or private entities in which recognition of the user is a key concern to the respective organization. SR, a goal of biometrics technology, are deploying increasingly in the secured organizations which is a task of identifying speaker using speaker behavioral/ physiological traits such as speech signal and visual signal which are carrying linguistic/non- linguistic information and visual information respectively. The development of SR system is still an active area of research which has a wide range of applications from national security, communication system security, computer security, computer network security, ecommerce to forensic (1-3). SR has been one of most important biometrics technologies during the past decades. It has a wide range of applications such as identity authentication, access control, and surveillance etc. From the beginning of 90's, new optimization technique researches using analogy of swarm behavior of natural creatures have been started. Dorigo developed ant colony optimization (ACO) mainly based on the social insect, especially ant, metaphor (4). Each individual exchanges information through pheromone implicitly in ACO. Eberhart and Kennedy developed particle swarm optimization (PSO) based on the analogy of swarm of bird and fish school (5). Each individual exchanges previous experiences in PSO. These researches are called Swarm Intelligence (4,5). In next section, we will introduce the Particle Optimization (PSO) method. Section III describes the experimental evaluation on speaker recognition. Finally, section IV summarizes the conclusion drawn from this study. II. Particle Optimization