Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
MISSING FEATURE IMPUTATION OF LOG-SPECTRAL DATA FOR NOISE ROBUST ASR Bengt J. Borgstrom and Abeer Alwan
 

Summary: MISSING FEATURE IMPUTATION OF LOG-SPECTRAL DATA FOR NOISE ROBUST ASR
Bengt J. Borgstrom and Abeer Alwan
Department of Electrical Engineering,
University of California, Los Angeles
jonas, alwan@ee.ucla.edu
ABSTRACT
In this paper, we present a missing feature (MF) imputation
algorithm for log-spectral data with applications to noise ro-
bust ASR. Drawing from previous work [1], we adapt the
previously proposed spectrographic reconstruction solution to
the liftered log-spectral domain by introducing log-spectral
flooring (LS-FLR). LS-FLR is shown to be an efficient and
effective noise robust feature extraction technique. When LS-
FLR is integrated in deriving the novel log-spectral data im-
putation framework, the overall system is shown to provide
significant improvements in noise robust speech recognition.
Index Terms-- Missing Features, Feature Extraction, Com-
pressibility, Noise Robust Automatic Speech Recognition.
1. INTRODUCTION
In-vehicle speech recognition remains a challenging task due

  

Source: Alwan, Abeer - Electrical Engineering Department, University of California at Los Angeles

 

Collections: Computer Technologies and Information Sciences