Speaker recognition with a MLP classifier and LPCC codebook

TL;DR

This paper enhances speaker recognition accuracy by combining MLP classifiers with LPCC codebooks and introduces an efficient algorithm to reduce computational complexity, achieving notable error rate reductions.

Contribution

It presents a novel linear combination approach for MLP and LPCC methods and an algorithm that significantly reduces LPCC-VQ system complexity.

Findings

Error rate reduced from 3.68% to 2.1%.

Recognition accuracy improved with combined methods.

Computational complexity decreased by a factor of 4.

Abstract

This paper improves the speaker recognition rates of a MLP classifier and LPCC codebook alone, using a linear combination between both methods. In simulations we have obtained an improvement of 4.7% over a LPCC codebook of 32 vectors and 1.5% for a codebook of 128 vectors (error rate drops from 3.68% to 2.1%). Also we propose an efficient algorithm that reduces the computational complexity of the LPCC-VQ system by a factor of 4.