Model based neuro-fuzzy ASR on Texas processor

TL;DR

This paper presents a speech recognition algorithm utilizing MFCC features and classifiers based on MLP and fuzzy inference, implemented on a 600 MHz Texas DSP, demonstrating high efficiency and performance.

Contribution

It introduces a novel neuro-fuzzy speech recognition system implemented on a fixed-point DSP, combining MLP and fuzzy inference classifiers.

Findings

High recognition accuracy demonstrated

Efficient implementation on fixed-point DSP

Effective use of neuro-fuzzy approach for ASR

Abstract

In this paper an algorithm for recognizing speech has been proposed. The recognized speech is used to execute related commands which use the MFCC and two kind of classifiers, first one uses MLP and second one uses fuzzy inference system as a classifier. The experimental results demonstrate the high gain and efficiency of the proposed algorithm. We have implemented this system based on graphical design and tested on a fix point digital signal processor (DSP) of 600 MHz, with reference DM6437-EVM of Texas instrument.