Fourier-Lapped Multilayer Perceptron Method for Speech Quality Assessment
- PDF / 405,516 Bytes
- 10 Pages / 600 x 792 pts Page_size
- 39 Downloads / 234 Views
Fourier-Lapped Multilayer Perceptron Method for Speech Quality Assessment ´ Vidal Ribeiro Moises Departamento de Comunicac¸o˜es (DECOM), Faculdade de Engenharia El´etrica e de Computac¸a˜ o (FEEC), Universidade Estadual de Campinas (UNICAMP), Caixa Postal 6101, 13083-852 Campinas SP, Brazil Email: [email protected]
Jayme Garcia Arnal Barbedo Departamento de Comunicac¸o˜es (DECOM), Faculdade de Engenharia El´etrica e de Computac¸a˜ o (FEEC), Universidade Estadual de Campinas (UNICAMP), Caixa Postal 6101, 13083-852 Campinas SP, Brazil Email: [email protected]
˜ Marcos Travassos Romano Joao Departamento de Comunicac¸o˜es (DECOM), Faculdade de Engenharia El´etrica e de Computac¸a˜ o (FEEC), Universidade Estadual de Campinas (UNICAMP), Caixa Postal 6101, 13083-852 Campinas SP, Brazil Email: [email protected]
Amauri Lopes Departamento de Comunicac¸o˜es (DECOM), Faculdade de Engenharia El´etrica e de Computac¸a˜ o (FEEC), Universidade Estadual de Campinas (UNICAMP), Caixa Postal 6101, 13083-852 Campinas SP, Brazil Email: [email protected] Received 1 November 2003; Revised 31 August 2004 The paper introduces a new objective method for speech quality assessment called Fourier-lapped multilayer perceptron (FLMLP). This method uses an overcomplete transform based on the discrete Fourier transform (DFT) and modulated lapped transform (MLT). This transform generates the DFT and the MLT speech spectral domains from which several relevant perceptual parameters are extracted. The proposed method also employs a multilayer perceptron neural network trained by a modified version of the scaled conjugated gradient method. This neural network maps the perceptual parameters into a subjective score. The numerical results show that FLMLP is an effective alternative to previous methods. As a result, it is worth stating that the techniques here described may be potentially useful to other researches facing the same kind of problem. Keywords and phrases: fast Fourier transform, modulated lapped transform, neural network, objective speech quality assessment, perceptual feature, scaled conjugated gradient optimization method.
1.
INTRODUCTION
The continuous search for efficient and reliable speech transmissions through communication channels has produced a great number of speech devices (specially codecs), which often include highly sophisticated features, making their quality assessment a tricky task. For many years, the assessment of speech devices has been mostly carried out using subjective tests, in which human listeners perform the evaluation. This kind of test, although very accurate, is quite expensive and timeconsuming. Such situation has motivated the search for objective methods able to suitably replace the subjective tests.
Several objective methods have been proposed [1, 2, 3, 4, 5, 6, 7, 8, 9, 10] so far. Among them, PESQ (perceptual evaluation of speech quality) [7], which is currently adopted as a standard by the International Telecommunication Union (ITU), aggregates some of the best fea
Data Loading...
