FPGA-Based real-time speech recognition system

Abstract

he main objective of this project is to design and implement a real time speech recognition system using a Field Programmable Gate Array (FPGA). Our system will be based on the Mel Frequency Cepstral Coefficients (MFCC) and a distance-based matching algorithm which fit very well the targeted system. The model will be implemented on the DE2 FPGA board including both SOPC and non-SOPC modules. Spoken words will be entered via a microphone connected to the audio input of the board. The recognized words will be displayed on the LCD module of the board. The obtained system can be used in several applications such as mobile robot control with voice command.