DMCA
Speech Enhancement, Gain, and Noise Spectrum Adaptation Using Approximate Bayesian Estimation
| Citations: | 3 - 0 self |
BibTeX
@MISC{Hao_speechenhancement,,
author = {Jiucang Hao and Hagai Attias and Srikantan Nagarajan and Te-won Lee and Terrence J. Sejnowski},
title = {Speech Enhancement, Gain, and Noise Spectrum Adaptation Using Approximate Bayesian Estimation},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract—This paper presents a new approximate Bayesian estimator for enhancing a noisy speech signal. The speech model is assumed to be a Gaussian mixture model (GMM) in the log-spectral domain. This is in contrast to most current models in frequency domain. Exact signal estimation is a computationally intractable problem. We derive three approximations to enhance the efficiency of signal estimation. The Gaussian approximation transforms the log-spectral domain GMM into the frequency domain using minimal Kullback–Leiber (KL)-divergency criterion. The frequency domain Laplace method computes the maximum a posteriori (MAP) estimator for the spectral amplitude. Correspondingly, the log-spectral domain Laplace method computes the MAP estimator for the log-spectral amplitude. Further, the gain and noise spectrum adaptation are implemented using the expectation–maximization (EM) algorithm within the GMM under Gaussian approximation. The proposed algorithms are evaluated by applying them to enhance the speeches corrupted by the speech-shaped noise (SSN). The experimental results demonstrate that the proposed algorithms offer improved signal-to-noise ratio, lower word recognition error rate, and less spectral distortion.
Keyphrases
speech enhancement frequency domain gaussian approximation exact signal estimation divergency criterion signal estimation log-spectral domain gmm map estimator speech-shaped noise log-spectral domain laplace method speech model log-spectral domain noisy speech signal intractable problem spectral amplitude new approximate bayesian estimator algorithm offer word recognition error rate signal-to-noise ratio frequency domain laplace method log-spectral amplitude spectral distortion minimal kullback leiber gaussian mixture model noise spectrum adaptation current model expectation maximization experimental result
