• Publications
  • Influence
Speckle filtering of synthetic aperture radar images : a review
Abstract Speckle, appearing in synthetic aperture radar (SAR) images as granular noise, is due to the interference of waves reflected from many elementary scatterers. Speckle in SAR imagesExpand
  • 466
  • 33
Template-Based Continuous Speech Recognition
TLDR
We extend the DTW framework with a flexible subword unit mechanism and a class sensitive distance measure-two components suggested by state-of-the-art HMM systems. Expand
  • 172
  • 12
  • PDF
Coping with disfluencies in spontaneous speech recognition: Acoustic detection and linguistic context manipulation
TLDR
We propose novel methods that aim at coping with the problems induced by three types of disfluencies, namely filled pauses, repeated words and sentence restarts. Expand
  • 35
  • 9
A Review of Signal Subspace Speech Enhancement and Its Application to Noise Robust Speech Recognition
TLDR
The objective of this paper is threefold: (1) to provide an extensive review of signal subspace speech enhancement, (2) to derive an upper bound for the performance of these techniques, and (3) to present a comprehensive study of the potential of subspace filtering to increase the robustness of automatic speech recognisers against stationary additive noise distortions. Expand
  • 128
  • 5
  • PDF
Model-based feature enhancement with uncertainty decoding for noise robust ASR
TLDR
In this paper, several techniques are proposed to incorporate the uncertainty of the clean speech estimate in the decoding process of the backend recogniser in the context of model-based feature enhancement (MBFE) for noise robust speech recognition. Expand
  • 59
  • 4
SPRAAK: an open source "SPeech recognition and automatic annotation kit"
SPRAAK is a new open source speech recognition package. It is derived from the HMM package that has been developed over the past 15 years at ESAT, KULeuven and which has been in use by a number ofExpand
  • 63
  • 2
  • PDF
Data driven example based continuous speech recognition
TLDR
The dominant acoustic modeling methodology based on Hidden Markov Models is known to have certain weaknesses. Expand
  • 70
  • 2
  • PDF
Improved feature decorrelation for HMM-based speech recognition
TLDR
In most HMM-based recognition systems, a mixture of diagonal covariance gaussians is used to model the observation density functions in the states. Expand
  • 19
  • 2
  • PDF
Confidence scoring based on backward language models
TLDR
In this paper we introduce the backward N-gram language model (LM) scores as a confidence measure in large vocabulary continuous speech recognition. Expand
  • 24
  • 2
  • PDF
Development and application of computer vision systems for use in livestock production
TLDR
This paper examines the feasibility of applying computer vision systems to improve health, welfare and efficiency in livestock production with a multi-disciplinary, interactive approach to develop optimal solutions. Expand
  • 44
  • 2