Vocal Tract Area Function Estimation Using Particle Swarm

Abstract

The problem of estimating the Vocal Tract Area Functions (VTAF) from speech signals arise in many applications. However, the existing methods for solving this problem were designed for small subset of speech sounds. This research presents a novel method for estimating VTAF from speech signal based on Particle Swarm Optimization (PSO). This method is general and can be applied for large classes of speech sounds. The proposed method was tested and the results positively verified the effectiveness of using the particle swarm approach in estimating VTAF.