PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Missing-Feature Reconstruction With a Bounded Nonlinear State-Space Model
Ulpu Remes, Kalle Palomäki, Tapani Raiko, Antti Honkela and Mikko Kurimo
IEEE Signal Processing Letters Volume 18, Number 10, pp. 563-566, 2011.

Abstract

Missing-feature reconstruction can improve speech recognition performance in unknown noisy environments. In this work, we examine using a nonlinear state-space model (NSSM) for missing-feature reconstruction and propose estimation with observed bounds to improve the NSSM performance. Evaluated in large-vocabulary continuous speech recognition task with babble and impulsive noise, using observed bounds in NSSM state estimation significantly improved the method performance.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Speech
Theory & Algorithms
ID Code:8942
Deposited By:Kalle Palomäki
Deposited On:21 February 2012