In most cases the increase in voice coil temperature is the major cause for speaker failure.Several techniques have been developed to keep voice coil at room temperature under high power inputs.It is a known fact that rise in temperature causes a rise in the total resistance of the voice coil.Providing negative impedance at the amplifier output to compensate for the increase in voice coil resistance is not a new technique.Here we have developed a thermal model to predict the voice coil temperature at different instants of time through computer simulation without actually measuring the voltage across the voice coil and hence calculate the negative impedance required to counteract the increase in voice coil resistance.

View full abstract

Perception of missing fundamental is widespread in vertebrate animals but seldom it is discussed with the relationship with vocal signals used by the animals.We tested the perception of missing fundamentals in two species of closely related finches with widely different vocal signals.Zebra finches and Bengalese finches were trained, in a Go-Nogo operant task, to discriminate between two harmonic complexes that were constructed by adding integer multiples of a 400Hz or a 652Hz.Both stimuli produced the perception of missing fundamentals to human listeners.After the birds learned the task, 4 sine waves, the fundamental frequency of each of the training stimuli and the geometric average frequency of each, were presented as probe stimuli along with the original training stimuli.Both species responded more to the missing fundamental of the Go stimuli than to that of Nogo stimuli.However, both species of birds responded more to the geometric averages of the stimuli than to the missing fundamentals.Thus, in these birds, the perception of the spectral pitch is dominant over the perception of the periodicity pitch and this tendency was not dependent upon the types of vocalizations each species produced.

View full abstract

The paper examines how well standard sources of impact sounds(bang-machine, A-ball, NF-ball, new rubber ball)imitate real impact sounds(human jumping and hopping).The compared sounds were described from psychoacoustic point of view.At the same time similarities between the impact sounds were judged in the psychoacoustic experiment.Results were compared with the existing standards expressed in F- and S-maximum sound levels:L_{A max, F}, L_{A max, S} and sound exposure level L_AE.

View full abstract

Context-dependent phone units, such as triphones, have recently come to be used to model subword units in speech recognition systems that are based on the use of hidden Markov models(HMMs).While most such systems employ clustering of the HMM parameters(e.g., subword clustering and state clustering)to control the HMM size, so as to avoid poor recognition accuracy due to a lack of training data, none of them provide any effective criteria for determining the optimal number of clusters.This paper proposes a method in which state clustering is accomplished by way of phonetic decision trees and in which the minimum description length(MDL)criterion is used to optimize the number of clusters.Large-vocabulary Japanese-language recognition experiments show that this method achieves higher accuracy than the maximum-likelihood approach.

View full abstract

A 27-node isoparametric acoustic element, namely Spl27, using the spline interpolation polynomials for the analysis of sound fields in rooms is presented first.Next, the basic characteristics of the element are discussed by comparing with the conventional elements in the eigenanalysis of a small room.Then, the mechanism of causing errors in eigenanalysis is investigated to propose error-characteristic curves as proper guidelines for the appropriate applications of the elements.An example application on a three dimensional sound field proved that, if the size of elements satisfies the value given by the guideline, Spl27 can be expected to provide both fair eigenfrequency approximation within 1% relative error and exact modal order agreement with the analytic solution.Finally, its application to a one-dimensional sound field proved that the guideline estimated by the error-characteristic curves give satisfactory results in the approximation of sound pressure waveforms.The basic relation between the eigenmode approximation and resulting sound pressure response was clarified through the process.

View full abstract

When using wine glasses as musical instruments(also referred to as the glass-harp), the pitch needs to be minutely adjusted.That is, it needs to be tuned.A wide adjustment range has been achieved by a new method that locally shaves the bottom of the cup of each vessel circumferencially.The pitch decreased in proportion to the quantity of glass shaved.This relationship between the quantity of glass shaved and the change in pitch was clarified both experimentally and analytically by Finite Element Method(FEM)analysis.The amount of pitch change accompanied with the shaving method is occasionally limited by the vessel shape.In such cases, pitch can be changed by filling wine glasses with specific quantities of water, a well-known conventional tuning method.This auxiliary method has been measured experimentally and analyzed by FEM to clarify the relationship between the water quantity in vessels and the amount of pitch change.A harmonics analysis was also performed.Using these procedures, prediction of vibration frequency could be done in advance, which means a desired pitch can be easily obtained.

View full abstract