Search

A simple adaptive voice activity detector (VAD) is implemented using Gabor and gammatone atomic decomposition of speech for high Gaussian noise environments. Matching pursuit is used for atomic decomposition, and is shown to achieve optimal speech detection capability at high data compression rates for low signal to noise ratios. The most active dictionary elements found by matching pursuit are used for the signal reconstruction so that the algorithm adapts to the individual speakers dominant time-frequency characteristics. Speech has a high peak to average ratio enabling matching pursuit greedy heuristic of highest inner products to isolate high energy speech components in high noise environments. Gabor and gammatone atoms are both investigated with identical logarithmically spaced center frequencies, and similar bandwidths. The algorithm performs equally well for both Gabor and gammatone atoms with no significant statistical differences. The algorithm achieves 70% accuracy at a 0 dB SNR, 90% accuracy at a 5 dB SNR and 98% accuracy at a 20dB SNR using 30d B SNR as a reference for voice activity., {"references":["Gabor, D., Theory of communication, J. Inst. Elect. Eng., 93, pp. 429–\n457. 1946","Lobo, A., Loizou, P., Voiced/unvoiced speech discrimination in noise\nusing Gabor atomic decomposition. ICASSP (1) 2003: 820-823","Smith, E., Lewicki, M., Efficient auditory coding. Nature,\n439(7079):978–82, 2006.","R. Patterson I. Nimmo-Smith. An Efficient Auditory Filterbank Based\non the Gammatone Function. Institute of Acoustics on Auditory\nModelling 1987","Slaney, M., (1998) \"Auditory Toolbox Version 2\", Technical Report\n#1998-010, Interval Research Corporation, 1998.","Atlas, L. Decomposition of speech and sound into Modulations and\nCarriers. http://msrvideo.vo.msecnd.net/rmcvideos/173320/dl/\n173320.pdf, Microsoft Research & University of Washington. 2012","Mallat, S., Zhang, Z., Matching Pursuits with Time-Frequency\nDictionaries. IEEE transactions on signal processing, Vol 41. No 12,\n1993","Kressner, A., Anderson, D., Rozell, C. Causal Binary Mask Estimation\nfor Speech Enhancements using Sparsity Constraints. Proceedings on\nMeetings on Acoustics Vol. 9, 055037 2013","Guo, D., Verdu', S., Mutual Information and Minimum Mean-Square\nError in Gaussian Channels. IEEE transactions on information theory,\nVol. 51, No. 4, 2005\n[10] Eargle, J., Handbook of Recording Engineering. 4th Addition. Springer\nScience and Business Media. ISBN 1-4020-7230-9 (HC), 2003."]}

A study was designed to ascertain the influence of in ovo site of inoculation and embryonic fluid type on the development of Marek's disease (MD) vaccine viremia and efficacy against MD challenge. The experiments were divided into in vitro and in vivo phases. In the in vitro phase, herpesvirus of turkeys/SB-1 vaccine was combined with basal medium eagle (BME) medium (control), amniotic fluid, or allantoic fluid and subsequently titrated on secondary chick embryo fibroblast cultures. There were no significant differences in titer between the virus inoculum carried in BME and the virus inoculum combined with either the allantoic fluid or the amniotic fluid. In the in vivo phase, five routes of inoculation, amniotic, intraembryonic, allantoic, air cell, and subcutaneous at hatch, were compared for generation of protection against virulent MD challenge. Comparisons were made in both specific-pathogen-free and commercial broiler embryos/chicks and, for the amniotic and allantoic routes, injection at either day 17 or day 18 of embryonation. Reisolation of the vaccine virus at day 3 of age was also done for all routes with the exception of the air cell route. Vaccine virus was recovered from all birds tested that were injected in ovo via the amniotic and intraembryonic routes and the subcutaneously at hatch route but was isolated only sporadically from birds inoculated via the allantoic route. Vaccination protective efficacy against virulent MD for all birds vaccinated in ovo via the amniotic or intraembryonic routes and birds vaccinated subcutaneously at hatch was over 90% regardless of day of in ovo injection or bird type. Protective efficacy for vaccines delivered in ovo by either the allantoic or the air cell routes was less than 50% regardless of day of injection or bird type. Therefore, in ovo MD vaccines must be injected either via the amniotic route or the intraembryonic route for optimal performance.

The objective of the present study was to evaluate the protective effect of humoral and cellular immunities on turkeys infected with turkey coronavirus (TCV). Two trials were conducted with two separate hatches of turkey poults. Turkeys were experimentally immunosuppressed with cyclosporin A (CsA) or cyclophosphamide (CY) and infected with TCV. Prior to infection, treatment with CsA selectively suppressed T cell activity as revealed by 2–3 fold decreased (p

This PhD thesis grew out of the attempt to understand the forms in which the Thom-Porteous formula for the Chow ring could be extended to other oriented cohomology theories and in particular to algebraic cobordism, the universal such theory. For a suitably generic morphism of vector bundles f:E->F over a scheme X of pure dimension, the Thom-Porteous formula expresses the fundamental class of a degeneracy locus as a determinant in the Chern classes of E and F. The core of the proof consists in dealing with the universal case of morphisms of bundles over the flag bundle Fl(V). In this setting, the degeneracy loci are the so-called Schubert varieties, whose classes in the Chow ring are given by the double Schubert polynomials. The same approach led Buch to prove similar results for the Grothendieck ring of locally coherent sheaves which, once it is made into a graded ring, represents another example of oriented cohomology theory. In this case the fundamental classes of Schubert varieties are described by means of the double Grothendieck polynomials. In order to generalise these results, I used the work of Hornbostel and Kiritchenko on the push-forward morphism for P^1-bundles in algebraic cobordism to give an explicit description of the push-forward cobordism classes of the Bott-Samelson resolutions in the algebraic cobordism of Fl(V). These schemes over Fl(V) play an imporant role in the classical theory since every Schubert variety is birationally isomorphic to, at least, one member of the family. It then follows that knowing these classes in CH(Fl(V)) and K^0(Fl(V)) automatically gives the fundamental classes of all Schubert varieties. In the attempt to obtain an extension of this result for a more general oriented cohomology theory, I have specialized the expression obtained in the algebraic cobordism ring off Fl(V) to connected K-theory, the universal theory satisfying birational invariance for smooth schemes. In this context I have been able to give a geometrical interpretation to the double beta-polynomials of Fomin and Kirillov: the push-forward classes of the Bott-Samelson resolutions can be obtained by evaluating the double beta-polynomials at the Chern roots of the bundles used to describe the corresponding Schubert variety, exactly as it was happening for the Chow ring and for the Grothendieck ring.

The purpose of the present study was to examine the antigenicity of turkey coronavirus (TCV) isolates from various geographic areas with antibodies to different viruses. Seventeen isolates of TCV were recovered from intestinal samples submitted to Animal Disease Diagnostic Laboratory, Purdue University, from turkey farms located in different geographic areas. The prototype TCV Minnesota isolate (TCV-ATCC) was obtained from the American Type Culture Collection. Intestinal sections were prepared from turkey embryos infected with different TCV isolates and reacted with polyclonal or monoclonal antibodies to TCV, infectious bronchitis virus (IBV), bovine coronavirus (BCV), transmissible gastroenteritis virus (TGEV), reovirus, rotavirus, adenovirus, or enterovirus in immunofluorescent antibody staining. All 18 TCV isolates have the same antigenic reactivity pattern with the same panel of antibodies. Positive reactivity was seen with polyclonal antibodies to the TCV Indiana isolate, the TCV Virginia isolate, TCV-ATCC, and the IBV Massachusetts strain as well as monoclonal antibodies to the TCV North Carolina isolate or the membrane protein of IBV. Antibodies to BCV or TGEV were not reactive with any of the TCV isolates. Reactivity of antibodies to unrelated virus, rotavirus, reovirus, adenovirus, or enterovirus with different TCV isolates was all negative, except positive response was seen between enterovirus antibody and a TCV western North Carolina isolate, suggesting coinfection of turkeys with TCV and enterovirus in that particular case. The results indicated that the TCV isolates from these geographic locations in the U.S. shared close antigenicity and were antigenically related to IBV.