The present study examines the difference between categorization and goodness ratings in Udmurt (Finno-Ugric language) using different sets of spectral attributes. The tendency to have two areas of /u/ vowels was observed in languages with unrounded non-front closed vowels during the TURVOTES project. Our study explores whether this can be due to the different acoustic attributes used in the id...

We introduce a new class of exponentially accurate filters for processing piecewise smooth spectral data. Our study is based on careful error decompositions, focusing on a rather precise balance between physical space localization and the usual moments condition. Exponential convergence is recovered by optimizing the order of the filter as an adaptive function of both the projection order, and ...

spectral statistics, mixed systems, bifurcations For systems that are neither fully integrable nor fully chaotic, bifurcations of periodic orbits give rise to semiclassically emergent singularities in the fluctuating part N f1 of the energy-level counting function. The bifurcations dominate the spectral moments M m (h) = 〈 (N f1)2 m 〉 in the limit h? 0. We show that M m (h) ~ constant / h m , a...

The temperature dependence of the spectral distributions of the glass, crystalline, and supercooled-liquid phases of glycerol is investigated by means of neutron inelastic scattering. The importance of anharmonic effects is quantified by the temperature dependence of reduced spectral frequency moments. The onset of relaxational ~i.e., stochastic, zero-frequency! motions in the supercooled liqui...

Latent variable models are widely used in industry and research, though the problem of estimating their parameters has remained challenging; standard techniques (e.g., Expectation-Maximization) offer weak guarantees of optimality. There is a growing body of work reducing latent variable estimation problems to a certain(orthogonal) spectral decompositions of symmetric tensors derived from the mo...

In this paper, a method for the digital simulation of wind velocity fields by Fractional Spectral Moment function is proposed. It is shown that by constructing a digital filter whose coefficients are the fractional spectral moments, it is possible to simulate samples of the target process as superposition of Riesz fractional derivatives of a Gaussian white noise processes. The key of this simul...

A transport theory which is not restricted to the gradient and quasi-particle approximations is presented which is formulated in terms of the energy moments, or equivalently the equal-time derivatives of the one-particle Green functions. A set of infinite hierarchies of kinetic and constraint equations for equal-time quantities for the spectral and the kinetic part of the one-particle Green fun...

This paper presents a new approach to estimate the spectral density matrix of a strictly stationary r-vector valued time series in the case where there are some randomly missing observations. The modified series is defined and is used to construct the finite Fourier transform in L-disjoint segments of observations. The modified periodogram is defined and the spectral density estimate is constru...

Hidden semi-Markov models (HSMMs) are latent variable models which allow latent state persistence and can be viewed as a generalization of the popular hidden Markov models (HMMs). In this paper, we introduce a novel spectral algorithm to perform inference in HSMMs. Unlike expectation maximization (EM), our approach correctly estimates the probability of given observation sequence based on a set...

For the determination of improved energy levels of ionic spectra of elements with large values of nuclear magnetic dipole moment (and eventually large values of nuclear quadrupole moments), it is necessary to determine the center of gravity of spectral lines from resolved hyperfine structure patterns appearing in highly resolved spectra. This is demonstrated on spectral lines of Ta II, Pr II an...