Robust Speaker Recognition in Noisy Conditions

This paper investigates the problem of speaker identi-fication and verification in noisy conditions, assuming that speechsignals are corrupted by environmental noise, but knowledgeabout the noise characteristics is not available. This research ismotivated in part by the potential application of speaker recog-nition technologies on handheld devices or the Internet. Whilethe technologies promise an additional biometric layer of securityto protect the user, the practical implementation of such systemsfaces many challenges. One of these is environmental noise. Due tothe mobile nature of such systems, the noise sources can be highlytime-varying and potentially unknown. This raises the require-ment for noise robustness in the absence of information about thenoise. This paper describes a method that combines multicondi-tion model training and missing-feature theory to model noisewith unknown temporal-spectral characteristics. Multiconditiontraining is conducted using simulated noisy data with limitednoise variation, providing a “coarse” compensation for the noise,and missing-feature theory is applied to refine the compensationby ignoring noise variation outside the given training conditions,thereby reducing the training and testing mismatch. This paperis focused on several issues relating to the implementation of thenew model for real-world applications. These include the gener-ation of multicondition training data to model noisy speech, thecombination of different training data to optimize the recognitionperformance, and the reduction of the model’s complexity. Thenew algorithm was tested using two databases with simulated andrealistic noisy speech data. The first database is a redevelopmentof the TIMIT database by rerecording the data in the presence ofvarious noise types, used to test the model for speaker identifica-tion with a focus on the varieties of noise. The second database isa handheld-device database collected in realistic noisy conditions,used to further validate the model for real-world speaker verifica-tion. The new model is compared to baseline systems and is foundto achieve lower error rates.

A Context-Dependent Algorithm for Merging Uncertain Information in Possibility Theory

The need to merge multiple sources of uncertaininformation is an important issue in many application areas,especially when there is potential for contradictions betweensources. Possibility theory offers a flexible framework to represent,and reason with, uncertain information, and there isa range of merging operators, such as the conjunctive anddisjunctive operators, for combining information. However, withthe proposals to date, the context of the information to be mergedis largely ignored during the process of selecting which mergingoperators to use. To address this shortcoming, in this paper,we propose an adaptive merging algorithm which selects largelypartially maximal consistent subsets (LPMCSs) of sources, thatcan be merged through relaxation of the conjunctive operator, byassessing the coherence of the information in each subset. In thisway, a fusion process can integrate both conjunctive and disjunctiveoperators in a more flexible manner and thereby be morecontext dependent. A comparison with related merging methodsshows how our algorithm can produce a more consensual result.

Comparison of three adjuvants used to produce polyclonal antibodies to veterinary drugs

Two commercially available adjuvants, Gerbu LQ 3000 and Montanide ISA 50V, were assessed as potential replacements for Freund's adjuvant by evaluating their efficacy in the production of polyclonal antibodies to veterinary drugs in rabbits. The aim was to find an adjuvant that could produce a similar (or enhanced) immune response in the host animal without the undesirable side effects associated with Freund's complete and incomplete adjuvant. The assessment involved the examination of each injection site and the characterisation of the resultant antibodies with regards to antibody titre and sensitivity. It was found that the rabbits immunised with Gerbu adjuvant produced some of the most sensitive antibodies. However, titres were relatively low and adverse effects at injection sites were relatively common. Montanide adjuvant produced no adverse effects and the related antibodies were found to be of adequate sensitivity when compared to those from rabbits immunised with Freund's. It was concluded that Montanide ISA 50V could be considered as a suitable replacement to Freund's for the production of polyclonal antibodies, to low molecular weight compounds in rabbits. (C) 2007 Elsevier B.V. All rights reserved.

Protection of cattle against a natural infection of Fasciola hepatica by vaccination with recombinant cathepsin L1 (rFhCL1)

The liver fluke, Fasciola hepatica causes liver fluke disease, or fasciolosis, in ruminants such as cattle and sheep. An effective vaccine against the helminth parasite is essential to reduce our reliance on anthelmintics, particularly in light of frequent reports of resistance to some frontline drugs. In our study, Friesian cattle (13 per group) were vaccinated with recombinant F. hepatica cathepsin L1 protease (rFhCL1) formulated in mineral-oil based adjuvants, Montanide (TM) ISA 70VG and ISA 206VG. Following vaccination the animals were exposed to fluke-contaminated pastures for 13 weeks. At slaughter, there was a significant reduction in fluke burden of 48.2% in the cattle in both vaccinated groups, relative to the control non-vaccinated group, at p

HPLC Determination of Angiotensin-Converting Enzyme Activity on Toyopearl HW-40S Column

Angiotensin-converting enzyme (EC3.4.15. I; ACE), isa membrane-bounddipeptidyl carboxypeptidase that mediates the cleavage of the C-terminal dipeptide His-Leu of the decapeptide angiotensin, generating the most powerful endogenous vaso-constricting angiotensin.
Some ACE inhibitors, such as Captopril, have been used as anti-hypertensive drugs. Moreover in recent years, large quantities of ACE inhibitors have been identijied and isolated from peptides derivedfrom food material such as casein, soy protein, jish protein and so on. Functional food with hypotensive effect has been developed on the basis of these works.
Typicalprocedures for screening hypotensive peptides offood origins are separationof products of peptic and tryptic digestion of proteins followed by inhibitory activitydetermination of each fraction. A method developed by Cushman has been the mostwidely used, in which ACE activity is determined by the amount of hippuric acid
generated as a product of enzymatic reaction of ACE with tripeptide of hippuryl-Lhistidyl-L-leucine. Hippuric acid is determined spectrophotometrically at 228 nm after its isolation from the reaction system by ethylacetate extraction, which not only requires alarge quantity of reagent but also results in large error.
An improved method based on Cushman ’s method is proposed in this paper. In this method, an enzymatic reaction system is based on Cushman’s method, while isolation and determination of hippuric acid is performed by medium perjormance gel chromatography on a Toyopearl HW-40s column. Due to the size exclusion nature of the column with somewhat hydrophobic properties, complete separation of four existing fractions in the reaction system is obtained within a smallfraction of the time necessary in Cushman’s method, with ideal reproducibility.