Voice trigger system using fuzzy logic

Abstract

In this paper, we propose a voice trigger system using a keyword-dependent speaker recognition technique. The voice trigger system must be able to perform keyword recognition, as well as speaker recognition, without using computationally demanding speech recognizer, to properly trigger a mobile device with low computational power consumption. This paper presents a fuzzy logic system to classify the speaker voice features. Experiments using a Hebrew word bank show that the fuzzy based method performs with low recognition error while highly minimizing the power and area footprint comparing to conventional methods.