R
Robert I. Damper
Researcher at University of Southampton
Publications - 243
Citations - 3555
Robert I. Damper is an academic researcher from University of Southampton. The author has contributed to research in topics: Speaker recognition & Pronunciation. The author has an hindex of 28, co-authored 241 publications receiving 3409 citations. Previous affiliations of Robert I. Damper include Nanyang Technological University & Imperial College London.
Papers
More filters
Journal ArticleDOI
Band Selection for Hyperspectral Image Classification Using Mutual Information
TL;DR: A new strategy is described to estimate the MI using a priori knowledge of the scene, reducing reliance on a "ground truth" reference map, by retaining bands with high associated MI values (subject to the so-called "complementary" conditions).
Journal ArticleDOI
Customizing Kernel Functions for SVM-Based Hyperspectral Image Classification
TL;DR: Spectrally weighted kernels based on learning weights by gradient descent are proposed and found to be slightly better than an alternative method based on estimating ";relevance"; between band information and ground truth.
Journal ArticleDOI
A multistrategy approach to improving pronunciation by analogy
TL;DR: This paper extends previous work on PbA in several directions, including full pattern matching between input letter string and dictionary entries, as well as including lexical stress in letter-to-phoneme conversion and extended the method to phoneme- to-letter conversion.
Proceedings Article
Testing an auditory model by resynthesis
R. W. Hukin,Robert I. Damper +1 more
TL;DR: It is shown that a spectral representation in which approximately two-thirds of the FFT frequency components are discarded, but the DOMIN representation is unchanged, can produce resynthesised speech of high intelligibility and is presented evidence showing that testing by resynthesis is superior to the alternative techniques for assessing auditory models.
Journal ArticleDOI
A fast separability-based feature-selection method for high-dimensional remotely sensed image classification
TL;DR: It is shown that there is a general framework based on the criterion of mutual information (MI) that can provide a realistic solution to the problem of feature selection for high-dimensional data and a fast feature-selection scheme based on a 'greedy' optimisation strategy is proposed.