VOICE EXTRACTION FOR SPEECH RECOGNITION BASED ON NEURAL NETWORK

Abstract

This paper presents the framework of malay isolated digit speech recognition system. The framework
design is based on neural network method. One of the very famous methods to develop speech recognition system is a
neural network (NN). NN is a computational paradigm model that consists of interconnected nerve cells. The NN is capable
to classify noisy data, various pattern data, variable data streams, multiple data and overlapping, interacting and
incomplete cues. It has been used for many different tasks because its capability in solving the non-linear problems. Thus,
the versatility of NN makes them significant for automatic speech recognition (ASR). In speech recognition, the speech
signal needs to be processed before being applied to neural network classifier and this process is known as preprocessing
phase. It is well known preprocessing that is one of the data trimming tools used in preprocessing phase. In this study we
will discuss fundamental design of preprocessing for speech recognition based on NN.