Extending the articulation index to account for non-linear distortions introduced by noise-suppression algorithms

The conventional articulation index (AI) measure cannot be applied in situations where non-linear operations are involved and additive noise is present. This is because the definitions of the target and masker signals become vague following non-linear processing, as both the target and masker signals are affected. The aim of the present work is to modify the basic form of the AI measure to account for non-linear processing. This was done using a new definition of the output or effective SNR obtained following non-linear processing. The proposed output SNR definition for a specific band was designed to handle cases where the non-linear processing affects predominantly the target signal rather than the masker signal. The proposed measure also takes into consideration the fact that the input SNR in a specific band cannot be improved following any form of non-linear processing. Overall, the proposed measure quantifies the proportion of input band SNR preserved or transmitted in each band after non-linear processing. High correlation (r?=?0.9) was obtained with the proposed measure when evaluated with intelligibility scores obtained by normal-hearing listeners in 72 noisy conditions involving noise-suppressed speech corrupted in four different real-world maskers.     

An articulation index based procedure for predicting the speech recognition performance of hearing-impaired individuals

An articulation index calculation procedure developed for use with individual normal-hearing listeners [C. Pavlovic and G. Studebaker, J. Acoust. Soc. Am. 75, 1606-1612 (1984)] was modified to account for the deterioration in suprathreshold speech processing produced by sensorineural hearing impairment. Data from four normal-hearing and four hearing-impaired subjects were used to relate the loss in hearing sensitivity to the deterioration in speech processing in quiet and in noise. The new procedure only requires hearing threshold measurements and consists of the following two modifications of the original AI procedure of Pavlovic and Studebaker (1984): The speech and noise spectrum densities are integrated over bandwidths which are, when expressed in decibels, larger than the critical bandwidths by 10% of the hearing loss. This is in contrast to the unmodified procedure where integration is performed over critical bandwidths. The contribution of each frequency to the AI is the product of its contribution in the unmodified AI procedure and a "speech desensitization factor." The desensitization factor is specified as a function of the hearing loss. The predictive accuracies of both the unmodified and the modified calculation procedures were assessed by comparing the expected and observed speech recognition scores of four hearing-impaired subjects under various conditions of speech filtering and noise masking. The modified procedure appears accurate for general applications. In contrast, the unmodified procedure appears accurate only for applications where results obtained under various conditions on a single listener are compared to each other.     

Application of the articulation index to the speech recognition of normal and impaired listeners wearing hearing protection

The present study examined the application of the articulation index (AI) as a predictor of the speech-recognition performance of normal and hearing-impaired listeners with and without hearing protection. The speech-recognition scores of 12 normal and 12 hearing-impaired subjects were measured for a wide range of conditions designed to be representative of those in the workplace. Conditions included testing in quiet, in two types of background noise (white versus speech spectrum), at three signal-to-noise ratios (+ 5, 0, - 5 dB), and in three conditions of protection (unprotected, earplugs, earmuffs). The mean results for all 21 listening conditions and both groups of subjects were accurately described by the AI. Moreover, a single transfer-function relating performance to the AI could describe all the data from both groups.     

Whispering gallery mode selective reflector with a high azimuthal index for the input cavity of a gyroklystron

A selective reflector in the form of a resonance extension is suggested for the input cavity of a 8-mm-wave gyroklystron operating at whispering gallery mode H _{m, 1} with an azimuthal index of several tens. Simulation data for the selective reflector are presented.     

The influence of the input beam width on the bandwidth of multimode step index fibers

A detailed theoretical investigation of the influence of the input on the bandwidth of multimode step index fibers is given. The calculations take into account mode coupling, absorption and leaky modes. For 100 m of a typical large numerical aperture fiber the bandwidth can vary with excitation from 1.8 GHz to 15 GHz. The numerical calculations agree with published experimental results.     

Acoustic properties for place of articulation in nasal consonants

The goal of this study was to determine whether acoustic properties could be derived for English labial and alveolar nasal consonants that remain stable across vowel contexts, speakers, and syllable positions. In experiment I, critical band analyses were conducted of five tokens each of [m] and [n] followed by the vowels [i e a o u] spoken by three speakers. Comparison of the nature of the changes in the spectral patterns from the murmur to the release showed that, for labials, there was a greater change in energy in the region of Bark 5-7 relative to that of Bark 11-14, whereas, for alveolars, there was a greater change in energy from the murmur to the release in the region of Bark 11-14 relative to that of Bark 5-7. Quantitative analyses of each token indicated that over 89% of the utterances could be appropriately classified for place of articulation by comparing the proportion of energy change in these spectral regions. In experiment II, the spectral patterns of labial and alveolar nasals produced in the context of [s] + nasal ([ m n]) + vowel ([ i e a o u]) by two speakers were explored. The same analysis procedures were used as in experiment I. Eighty-four percent of the utterances were appropriately classified, although labial consonants were less consistently classified than in experiment I. The properties associated with nasal place of articulation found in this study are discussed in relation to those associated with place of articulation in stop consonants and are considered from the viewpoint of a more general theory of acoustic invariance.     

Relational spectral features for place of articulation in nasal consonants

Much recent research on acoustic cues for consonants' places of articulation has focused upon the nature of the rapid spectral changes that take place between signal portions corresponding to consonantal closure and adjacent vowels. The study reported here builds on the foundation laid by earlier studies that have explored techniques for representing spectral change and for classifying place of articulation of nasal consonants using features extracted from rapid spectral changes that take place over murmur-to-vowel transitions. A new procedure is reported that avoids the use of predetermined absolute frequency bands in deriving parameters of spectral change in nasals. In experiments using the speech of 20 female and 20 male talkers, in a variety of physical and perceptual spectral scalings, application of the new procedure results in 77% correct classification of place of articulation of syllable-initial nasals and 51% correct classification of place of articulation of syllable-final nasals (for which there is a three-way contrast). Tested on the same data, a technique using predetermined absolute frequency bands produced 72% correct classification of syllable-initial nasals.     

Solutions for the multiple input/output poblem

This paper provides new solutions for the general multiple input/output problem involving arbitrary stationary random processes by using a special representation for random records. Mathematical results are obtained here for partial coherence functions, multiple coherence functions, decomposition of output spectra into physically meaningful components, and other useful quantities. These results are relatively simple to understand and to compute compared to previously known least-squares methods. Engineering applications and error analysis matters are also considered.     

Synergistic modes of vocal tract articulation for American English vowels

The purpose of this study was to investigate the spatial similarity of vocal tract shaping patterns across speakers and the similarity of their acoustic effects. Vocal tract area functions for 11 American English vowels were obtained from six speakers, three female and three male, using magnetic resonance imaging (MRI). Each speaker's set of area functions was then decomposed into mean area vectors and representative modes (eigenvectors) using principal components analysis (PCA). Three modes accounted for more than 90% of the variance in the original data sets for each speaker. The general shapes of the first two modes were found to be highly correlated across all six speakers. To demonstrate the acoustic effects of each mode, both in isolation and combined, a mapping between the mode scaling coefficients and [F1, F2] pairs was generated for each speaker. The mappings were unique for all six speakers in terms of the exact shape of the [F1, F2] vowel space, but the general effect of the modes was the same in each case. The results support the idea that the modes provide a common system for perturbing a unique underlying neutral vocal tract shape.     

Perceptual integration of the murmur and formant transitions for place of articulation in nasal consonants

This study reassessed the role of the nasal murmur and formant transitions as perceptual cues for place of articulation in nasal consonants across a number of vowel environments. Five types of computer-edited stimuli were generated from natural utterances consisting of [m n] followed by [i e a o u]: (1) full murmurs; (2) transitions plus vowel segments; (3) the last six pulses of the murmur; (4) the six pulses starting from the beginning of the formant transitions; and (5) the six pulses surrounding the nasal release (three pulses before and three pulses after). Results showed that the murmur provided as much information for the perception of place of articulation as did the transitions. Moreover, the highest performance scores for place of articulation were obtained in the six-pulse condition containing both murmur and transition information. The data support the view that it is the combination of nasal murmur plus formant transitions which forms an integrated property for the perception of place of articulation.     

The periodic system for free diatomic molecules—III. Theoretical articulation

A periodic system for free neutral diatomic molecules, previously proposed on the basis of tabulated data, is shown related topologically to the chart of the atoms. The relation is generalized to ionized atoms and ionized molecules. Support is drawn from tabulated data.     

Using auditory-visual speech to probe the basis of noise-impaired consonant-vowel perception in dyslexia and auditory neuropathy

Both dyslexics and auditory neuropathy (AN) subjects show inferior consonant-vowel (CV) perception in noise, relative to controls. To better understand these impairments, natural acoustic speech stimuli that were masked in speech-shaped noise at various intensities were presented to dyslexic, AN, and control subjects either in isolation or accompanied by visual articulatory cues. AN subjects were expected to benefit from the pairing of visual articulatory cues and auditory CV stimuli, provided that their speech perception impairment reflects a relatively peripheral auditory disorder. Assuming that dyslexia reflects a general impairment of speech processing rather than a disorder of audition, dyslexics were not expected to similarly benefit from an introduction of visual articulatory cues. The results revealed an increased effect of noise masking on the perception of isolated acoustic stimuli by both dyslexic and AN subjects. More importantly, dyslexics showed less effective use of visual articulatory cues in identifying masked speech stimuli and lower visual baseline performance relative to AN subjects and controls. Last, a significant positive correlation was found between reading ability and the ameliorating effect of visual articulatory cues on speech perception in noise. These results suggest that some reading impairments may stem from a central deficit of speech processing.     

Quantum filtering for multiple input multiple output systems driven by arbitrary zero-mean jointly Gaussian input fields

In this paper, we treat the quantum filtering problem for multiple input multiple output (MIMO) Markovian open quantum systems coupled to multiple boson fields in an arbitrary zero-mean jointly Gaussian state, using the reference probability approach formulated by Bouten and van Handel as a quantum version of a well-known method of the same name from classical nonlinear filtering theory, and exploiting the generalized Araki-Woods representation of Gough. This includes Gaussian field states such as vacuum, squeezed vacuum, thermal, and squeezed thermal states as special cases. The contribution is a derivation of the general quantum filtering equation (or stochastic master equation as they are known in the quantum optics community) in the full MIMO setup for any zero-mean jointly Gaussian input field states, up to some mild rank assumptions on certain matrices relating to the measurement vector.     

The effects of articulation on the acoustical structure of feline vocalizations

Feline isolation calls were analyzed, and a model was developed to relate the acoustical features of these calls to the physical processes used in their production. Fifty isolation calls were recorded from each of five cats for a total sample of 250 vocalizations. By combinations of Fourier transform, autocorrelation, and linear prediction methods, the fundamental frequency (glottal-pulse period) F0, the energy of F0, the frequency having maximum energy Fmax (not always F0), and the energy at this frequency were computed. Mean F0 ranged from 400-600 Hz for individual cats. For some cats F0 was consistent within calls, but for other cats sudden shifts in F0 occurred within calls. Here, Fmax was almost a harmonic of F0 and generally ranged from 1-2 kHz. For individual cats, the energy ratio E = (energy of Fmax/energy of F0) varied from 1 to 60 and the grand average E over the time course of the call varied from about 12 to 38. The mean rms call intensity was an inverted-U function of time. Measured jaw opening was strongly correlated with acoustical features of call. A Bessel-horn model with time-varying flare gave a good account of acoustical parameters such as Fmax. The presence of formantlike resonances in cat vocalizations and the important role of jaw movements (vocal gestures) in the production of these calls suggest that cats may provide a useful model for some aspects of human vocal behavior.     

On the refractive index for neutrons

The commonly used result for the neutron refractive index does not include effects due to binding of the nuclei or multiple scattering. We show how these effects can be incorporated in the theory of coherent neutron scattering and present an expression for the refractive index, which adds to the familiar result correction terms quadratic in the scattering length. Examples discussed in more detail include a calculation of the refractive indexn for H₂ gas. Corrections to 1 —n are of the order 10^–3. Results from our theory, which takes into account the actual dynamics of the scatterers, can differ substantially from those of the static approximation, in which the nuclei are held at fixed positions in space.     

Evaluating the Effective Action

The use of effective field theory, in situations wherein the energy-momentum of light particles is much lower than the rest mass of heavy degrees of freedom, has become an important one in contemporary physics. Herein we examine various means by which this effective action can be evaluated, using the effective photon–photon interaction—the Euler–Heisenberg Lagrangian—as a pedagogical example.