Harmonic extraction methods for the multiparameter eigenvalue problem willbe presented. These techniques are generalizations of their counterparts forthe standard and generalized eigenvalue problem. The methods aim to ap-proximate interior eigenpairs, generally more accurately than the standardextraction does. The process can be combined with any subspace expansionapproa...

In noisy listening conditions, the information available on which to base speech recognition decisions is necessarily incomplete: some spectro-temporal regions are dominated by other sources. We report on the application of a variety of techniques for missing data in speech recognition. These techniques may be based on marginal distributions or on reconstruction of missing parts of the spectrum...

Unvoiced speech separation is an important and challenging problem that has not received much attention. We propose a CASA based approach to segregate unvoiced speech from nonspeech interference. As unvoiced speech does not contain periodic signals, we first remove the periodic portions of a mixture including voiced speech. With periodic components removed, the remaining interference becomes mo...

The enhancement and detection of elongated structures in noisy image data is relevant for many biomedical imaging applications. To handle complex crossing structures in 2D images, 2D orientation scores U : R×S → C were introduced, which already showed their use in a variety of applications. Here we extend this work to 3D orientation scores U : R × S → C. First, we construct the orientation scor...

We propose a new cocktail-party recognition technique based on the coupling of a CASA-labelling method using the TDOA (Time Delay Of Arrival) with multistream recognition. This is an alternative to the classical "segregate and recognise" architecture. First, we have recorded a stereo database ST-NB95 from the mono Numbers95. This is composed of binary mixtures of sentences at 0dB, placed left a...

The purpose of this paper is to discuss the current partial solutions to the goal of semantic interoperability on the web. It is obvious that a general solution to the problem on a global level has not yet been achieved. We critically discuss the existing approaches, including technologies such as RDF, SHOE, PROMPT, Chimaera and others, trying to identify the most effective elements of each and...

Sperm with reduced motility has reduced fertilizing capacity. If low-motile Duroc sperm is mixed with high-motile sperm, the fertilizing capacity of that dose is equal to semen doses exclusively containing high-motile Duroc sperm. With doses that only contained low-motile sperm, the number of total born piglets per litter dropped by minimum 0.5. Duroc boars were selected on the basis of the per...

This paper outlines our recent research that is providing users with a 3D avatar representation, and in particular focuses on studies in which the avatar is self-animated in real time. We use full body motion tracking, so when participants move their hands and feet, these movements are mapped onto the avatar. In a recent study (Dodds et al., CASA 2010), we found that a selfanimated avatar aided...

Speech segregation, or the cocktail party problem, has proven to be extremely challenging. This presentation describes a computational auditory scene analysis (CASA) approach to the cocktail party problem. This approach performs auditory segmentation and grouping in a two-dimensional time-frequency representation that encodes proximity in frequency and time, periodicity, amplitude modulation, a...

We propose a running demonstration of coupling between an intermediate processing step (named CASA), based on the harmonicity cue, and partial recognition, implemented with a HMM/ANN multistream technique [2]. The model is able to recognise words corrupted with narrow band noise, either stationary or having variable center frequency. The principle is to identify frame by frame the most noisy su...