• Publications
  • Influence
Directional processing of color images: theory and experimental results
The processing of color image data using directional information is studied. The class of vector directional filters (VDF), which was introduced by the authors in a previous work, is further
Generalized multichannel image-filtering structures
A new filter structure, the directional-distance filters (DDF), is introduced, which combine both VDF and VMF in a novel way and are shown to be robust signal estimators under various noise distributions and compare favorably to other multichannel image processing filters.
Improved speech-to-text translation with the Fisher and Callhome Spanish-English speech translation corpus
The Fisher and Callhome Spanish-English Speech Translation Corpus is introduced, supplementing existing LDC audio and transcripts with ASR 1-best, lattice, and oracle output produced by the Kaldi recognition system and English translations obtained on Amazon’s Mechanical Turk.
Generative Content Models for Structural Analysis of Medical Abstracts
It is demonstrated that Hidden Markov Models are capable of accurately capturing the structure of medical abstracts, and can achieve classification accuracy comparable to that of discriminative techniques.
Combining vector median and vector directional filters: the directional-distance filters
A new class of filters, the directional-distance filters (DDF), which combine both VDF and VMF in a novel way are introduced, which can eliminate the noise much more effectively than the VMF (even the impulsive noise), and that they have the property of chromaticity preservation.
Neural-Network Lexical Translation for Cross-lingual IR from Text and Speech
We propose a neural network model to estimate word translation probabilities for Cross-Lingual Information Retrieval (CLIR). The model estimates better probabilities for word translations than
Score normalization and system combination for improved keyword spotting
Two techniques are shown to yield improved Keyword Spotting (KWS) performance when using the ATWV/MTWV performance measures, which resulted in the highest performance for the official surprise language evaluation for the IARPA-funded Babel project in April 2013.
A relationship between quantization and watermarking rates in the presence of additive Gaussian attacks
A system which embeds watermarks in n-dimensional Gaussian data and distributes them in compressed form is studied, and two surprising facts are demonstrated.
Speech recognitionwith segmental conditional random fields: A summary of the JHU CLSP 2010 Summer Workshop
This paper summarizes the 2010 CLSP Summer Workshop on speech recognition at Johns Hopkins University, to improve on state-of-the-art speech recognition systems by using Segmental Conditional Random Fields (SCRFs) to integrate multiple types of information.
Morphological Segmentation for Keyword Spotting
This paper augments a state-of-the-art KWS system with sub-word units derived from supervised and unsupervised morphological segmentations, and compares with phonetic and syllabic segmentations.