FPGA Implementation of a Pipelined Gaussian Calculation for HMM-Based Large Vocabulary Speech Recognition

A scalable large vocabulary, speaker independent speech recognition system is being developed using Hidden Markov Models (HMMs) for acoustic modeling and a Weighted Finite State Transducer (WFST) to compile sentence, word, and phoneme models. The system comprises a software backend search and an FPGA-based Gaussian calculation which are covered here. In this paper, we present an efficient pipelined design implemented both as an embedded peripheral and as a scalable, parallel hardware accelerator. Both architectures have been implemented on an Alpha Data XRC-5T1, reconfigurable computer housing a Virtex 5 SX95T FPGA. The core has been tested and is capable of calculating a full set of Gaussian results from 3825 acoustic models in 9.03 ms which coupled with a backend search of 5000 words has provided an accuracy of over 80%. Parallel implementations have been designed with up to 32 cores and have been successfully implemented with a clock frequency of 133?MHz.

A comparison of nonstaggered compact FDTD schemes for the 3D wave equation

This paper aims at providing a better insight into the 3D approximations of the wave equation using compact finite-difference time-domain (FDTD) schemes in the context of room acoustic simulations. A general family of 3D compact explicit and implicit schemes based on a nonstaggered rectilinear grid is analyzed in terms of stability, numerical error, and accuracy. Various special cases are compared and the most accurate explicit and implicit schemes are identified. Further considerations presented in the paper include the direct relationship with other numerical approaches found in the literature on room acoustic modeling such as the 3D digital waveguide mesh and Yee's staggered grid technique.

Dust-acoustic wave modulation in the presence of superthermal ions

A study is presented of the nonlinear self-modulation of low-frequency electrostatic (dust acoustic) waves propagating in a dusty plasma, in the presence of a superthermal ion (and Maxwellian electron) background. A kappa-type superthermal distribution is assumed for the ion component, accounting for an arbitrary deviation from Maxwellian equilibrium, parametrized via a real parameter kappa. The ordinary Maxwellian-background case is recovered for kappa ->infinity. By employing a multiple scales technique, a nonlinear Schrodinger-type equation (NLSE) is derived for the electric potential wave amplitude. Both dispersion and nonlinearity coefficients of the NLSE are explicit functions of the carrier wavenumber and of relevant physical parameters (background species density and temperature, as well as nonthermality, via kappa). The influence of plasma background superthermality on the growth rate of the modulational instability is discussed. The superthermal feature appears to control the occurrence of modulational instability, since the instability window is strongly modified. Localized wavepackets in the form of either bright-or dark-type envelope solitons, modeling envelope pulses or electric potential holes (voids), respectively, may occur. A parametric investigation indicates that the structural characteristics of these envelope excitations (width, amplitude) are affected by superthermality, as well as by relevant plasma parameters (dust concentration, ion temperature).

Arbitrary amplitude ion-acoustic solitary excitations in the presence of excess superthermal electrons

Velocity distribution functions with an excess of superthermal particles are commonly observed in space plasmas, and are effectively modeled by a kappa distribution. They are also found in some laboratory experiments. In this paper we obtain existence conditions for and some characteristics of ion-acoustic solitary waves in a plasma composed of cold ions and kappa-distributed electrons, where kappa>3/2 represents the spectral index. As is the case for the usual Maxwell-Boltzmann electrons, only positive potential solitons are found, and, as expected, in the limit of large kappa one recovers the usual range of possible soliton Mach numbers, viz., 1 < M < 1.58. For lower values of kappa, modeling the presence of a greater superthermal component, the range of accessible Mach numbers is reduced. It is found that the amplitude of the largest possible solitons that may be generated in a given plasma (corresponding to the highest allowed Mach number for the given plasma composition) falls off with decreasing kappa, i.e., an increasing superthermal component. On the other hand, at fixed Mach number, both soliton amplitude and profile steepness increase as kappa is decreased. These changes are seen to be important particularly for kappa < 4, i.e., when the electrons have a "hard" spectrum.

Perceptual learning of acoustic noise generates memory-evoked potentials

Experience continuously imprints on the brain at all stages of life. The traces it leaves behind can produce perceptual learning [1], which drives adaptive behavior to previously encountered stimuli. Recently, it has been shown that even random noise, a type of sound devoid of acoustic structure, can trigger fast and robust perceptual learning after repeated exposure [2]. Here, by combining psychophysics, electroencephalography (EEG), and modeling, we show that the perceptual learning of noise is associated with evoked potentials, without any salient physical discontinuity or obvious acoustic landmark in the sound. Rather, the potentials appeared whenever a memory trace was observed behaviorally. Such memory-evoked potentials were characterized by early latencies and auditory topographies, consistent with a sensory origin. Furthermore, they were generated even on conditions of diverted attention. The EEG waveforms could be modeled as standard evoked responses to auditory events (N1-P2) [3], triggered by idiosyncratic perceptual features acquired through learning. Thus, we argue that the learning of noise is accompanied by the rapid formation of sharp neural selectivity to arbitrary and complex acoustic patterns, within sensory regions. Such a mechanism bridges the gap between the short-term and longer-term plasticity observed in the learning of noise [2, 4-6]. It could also be key to the processing of natural sounds within auditory cortices [7], suggesting that the neural code for sound source identification will be shaped by experience as well as by acoustics.