Evaluation of Sparsification algorithm and Its Application in Speaker Recognition System

摘要

This paper proposes spectral domain compression of the speech signal using novel sparsing algorithms. In a sparse algorithm, representation little quantity of coefficients holds a large proportion of the energy. Automatic Speaker Recognition (ASR) sparsity can play a major role to resolve big data issues in speech compression and its storage in the database, where the speech signal can be compressed before applying to ASR system and later can be used in speaker recognition. The Speech signal is converted to a spectral domain using Discrete Rajan Transform (DRT) and only first spectrum component has been retained forcing the remaining component to zero. The speech signal spectrum can be maximally compressed 8:1 ratio to the unique one. Spectrally compressed speech signal can be stored in the database and during training and testing time it can be synthesized using Inverse Discrete Rajan Transform (IDRT) in automatic speaker recognition. Acceptable speech signal spectral compression is 75% with Percentage of Identification Accuracy (PIA) of the speaker recognition system with sparsing is 95.3% and without sparsification 98.8% for TIMIT database respectively. In this paper the Novel Fuzzy Vector Quantization (NFVQ) feature matching technique was used, due to high accuracy.