Mobility

Wavelet Packet and Modular Arithmetic for Text-Independent Speaker Identification

Free registration required

Executive Summary

In this paper, the techniques of Wavelet Transform (WT) with modular arithmetic and neural network were developed for speech based text-independent speaker identification. The number of repeating the remainder applied for a speech signal with Wavelet Packet (WP) upon level three features extraction method was developed. 266 features fed to Probabilistic Neural Network (PNN) for classification. The functions of features extraction and classification are performed using the Modular arithmetic, Wavelet Packet and neural Networks (MWPPN) expert system.

  • Format: PDF
  • Size: 177.75 KB