Showing papers in "Canadian Acoustics in 2012"

Analysis of tongue shapes during the production of kannada consonants

Abstract: A study was conducted to focus on the relative displacement of the tongue in the production of retroflex consonants in Kannada language, calculated as the difference between the tongue shapes during the consonant closure and the rest position. It was found that the retroflex stop and the palatal affricate were characterized by greater displacement compared to the dental and velar stops, suggestive of the greater articulatory complexity of the former two. The two curves are considered significantly different in a given region if their 95% confidence intervals do not overlap. The results of the statistical comparison of tongue shapes for Kannada lingual consonants are parallel to the earlier findings of displacement differences based on the same data.

The association between speaker-dependent formant space estimates and perceived vowel quality

Abstract: Differences in speaker-dependent formant space estimates is discussed in terms of differing FF-scaling estimates. In the experiment, listeners were first trained to report apparent speaker FF-scaling using the training method outlined in Barreda & Nearey. 25 native speakers of Canadian English from the University of Alberta were drawn from a participant pool in which undergraduate linguistics students take part in experiments in exchange for partial course credit. During the testing phase, listeners were presented with fully-randomized, isolated-vowel stimuli. 18% of the variance in reported FF-scaling is found, with F1 accounting for 67.8%, 10 accounting for 28.1%, and F3 accounting for only 0.2% of the explained variance. Results show that FF-scaling has a significant negative effect on vowel openness, showing that for a given vowel sound, when listeners reported a higher FF-scaling, they were less likely to hear an open vowel.

Predicting accentedness: Acoustic measurements of chinese-accented english

Abstract: Listeners perceive foreign-accented speech as different from native speech because it deviates from native speaker acoustic targets. These deviations may occur across many acoustic dimensions such as word duration, vowel duration, vowel formant values, and voice onset time. Accentedness ratings were modeled using ordinary least squares linear regression performed in R using the rms package. Prior to analysis, formant values were log transformed and plotted revealing considerable variation in vowel space. The vowel-to-word ratio was calculated by dividing the vowel duration of a word by the total duration of that word. If accentedness is a result of non-native productions approaching to varying degrees native-like acoustic targets, quantifying the distance from the native speaker norm allows examination of how variation along different variables affects perceived accentedness.

Effects of exposure to loud noise on the hearing of the residents of calabar, nigeria

Abstract: The article examines the effects of exposure to loud noise on the hearing of the residents of Calabar, Nigeria. To determine how related the subjective responses, assessed by the use of questionnaires as the study instrument were to the objective responses measured with the sound level meter, the coefficients of correlation were calculated for the noise measurements. Noise is a disturbance to the human environment that is escalating at such a high rate that it will become a major threat to the quality of human lives if nothing is done to reduce it. Noise has been a constant threat since the industrial revolution. Too much noise exposure may cause a temporarily change in hearing or a temporarily ringing in your ears (tinnitus). These short-term problems usually go away within a few minutes or hours after leaving the noise. However, repeated exposures to loud noise can lead to permanent, incurable hearing loss or tinnitus.

Auv localization in an underwater acoustic positioning system

Abstract: The University of Victoria is designing an Integrated Acoustic System (IAS), a high-precision underwater acoustic positioning system, similar in operation to a commercial long baseline positioning unit The IAS will be capable of localizing a target within the lAS range to a sufficient accuracy for use as ground truth for testing onboard navigation systems The non-linear localization accuracy for a target is estimated and the target positional uncertainty is mapped as a function of position within the range by estimating the linearized posterior uncertainties of the source-location Results show excellent agreement with the true locations, and linearization error is seen to be small Since linearization errors are shown to be small for the test cases investigated above, linearized uncertainty estimates can now be used to estimate target positional uncertainty for locations throughout the range with a high level of confidence

Acoustic metrics for classroom performance- A literature review

Abstract: A literature review undertaken by an undergraduate research project to enumerate all possible metrics that can be used to assess completely the classroom performance from an acoustic perspective, evaluating the sufficiency of the three conventional metrics, is presented. Barron and Lee concluded that reverberation time determines early decay time, early-to-late sound index, and total sound-pressure level with volume. Bradley states that there are numerous predicators of speech intelligibly that must be considered as well. Neuman and Hochberg state that studies involving classrooms have focused on determining optimal reverberation through testing adults as opposed to children. Palovic points out that it is often difficult make direct comparisons between results, because different researchers use different parameters when dealing with speech intelligibility.

Llnguopalatal contact differences between Japanese geminate and singleton stops

Abstract: Japanese is known to exhibit a phonemic contrast between geminate and singleton consonants. This contrast has received considerable attention in phonetic literature, with most studies focusing on its most salient, durational properties - differences between geminated and singletons in the duration of the consonant and the preceding or following vowel. The results of this preliminary study confirm the hypothesis that durational differences in the production of Japanese geminate/singleton stops can be accompanied by differences in the degree of contact. The longer geminates tend to be articulated with a tighter constriction, compared to the shorter and weaker-articulated singletons. To conclude, the current study presented some evidence for the role of degree of contact in the realization of the Japanese geminate and singleton stops, contributing to the growing body of work on non-durational acoustic and articulatory correlates of the contrast.

Acoustic transmission loss measurements in queen charlotte Basin

Abstract: A transmission loss experiment was carried out during the winter in south Hecate Strait using a small airgun array source. Airgun pulses were recorded at horizontal receiver ranges between 20 m and 10 km using a bottom-mounted hydrophone recorder. Transmission loss values were computed by subtracting measured source levels from received sound levels in 1/3-octave bands. Transmission loss data were compared to predictions from a parabolic-equation (PE) sound propagation model coupled with an airgun array source level model. The measured transmission loss was characteristic of cylindrical spreading, with very little additional loss attributable to non-geometric terms. Mid-frequency (100-400 Hz) sound propagation was found to be best supported by the environment. The PE model predictions showed good agreement with the experimental data.

On the influence of the material properties of the external ear on occlusion effect simulations

Abstract: A two-level fractional factorial design was implemented to examine how the material properties of the external ear tissues influence numerical predictions of the occlusion effect (OE). A simplified 2D model was developed and successfully compared to an equivalent 3D model whose complex external ear geometry was reconstructed using 135 anatomical images of a female cadaver head. Outer circumferential boundaries of the skin and cartilage domains are fixed. Analysis of variance indicates significant single factor effects for skin, soft, and bony tissues. Skin, soft, and bony material properties are found to contribute significantly to simulated OEs. Mostly Poisson's ratio and Young's modulus of the skin tissue tend to exhibit effect estimates which are large enough to cause relevant variations in simulated OE data.

Acoustic localization of an unknown number of sources in an uncertain ocean environment

Abstract: This paper develops a new approach to simultaneous localization of an unknown number of ocean acoustic sources when properties of the environment are poorly known, based on minimizing the Bayesian information criterion (BIC) over source and environmental parameters. A Bayesian formulation is developed in which water-column and seabed parameters, noise statistics, and the number, locations, and complex strengths (amplitudes and phases) of multiple sources are considered unknown random variables constrained by acoustic data and prior information. The BIC, which balances data misfit with a penalty for extraneous parameters, is minimized using hybrid optimization (adaptive simplex simulated annealing) over environmental parameters and Gibbs sampling over source locations. Closed- form maximum-likelihood expressions for source strength and noise variance at each frequency allow these parameters to be sampled implicitly, substantially reducing the dimensionality of the inversion. Gibbs sampling and the implicit formulation provide an efficient scheme for adding and deleting sources during the optimization. A simulated example is presented which considers localizing a quiet submerged source in the presence of two loud interfering sources in a poorly-known shallow-water environment.

Modelling uncertainty in an underwater acoustic positioning system

Abstract: This paper develops a ray-based travel-time inversion to simulate the accuracy of an active underwater acoustic localization system, and examines the localization accuracy as a function of various sources of error and geometric and environmental factors. The system considered here simulates localizing an autonomous underwater vehicle using arrival times of acoustic transmissions from an onboard source as measured at hydrophones distributed spatially over a test range. Since localization uncertainty is a function of source location, uncertainties are calculated for the source at a grid of locations over the areas of the test bed. Localization accuracy is considered as a function of timing errors, uncertainty in hydrophone locations, target depth, variations in sound-speed profile, and hydrophone geometry.

Auditory spatial attention in a complex acoustic environment while walking: Investigation of dual-task performance

Abstract: Auditory spatial attention in a complex acoustic environment while walking was investigated to find the dual-task performance. Three undergraduate students, ages 19 to 26 years, with normal pure-tone air-conducted hearing thresholds for frequencies from 0.25 to 8 kHz, performed a word identification task in two experimental conditions. Testing was conducted in StreetLab in the Challenging Environment Assessment Laboratory (CEAL) at the Toronto Rehabilitation Institute. The stimuli for the listening task were all sentences recorded by four male talkers for the Coordinated Response Measure (CRM). All participants completed 8 sessions in each of two conditions; standing and walking. The standing and walking conditions differed in terms of whether or not there was a secondary task during listening. The high degree of similarity in listening task performance between the standing and walking conditions suggest that the listening abilities in multi-talker environments of these participants were not affected by the inclusion of a walking component.

Sparse broadband acoustic response estimation via a gaussian mixture model with application to m-ary orthogonal signaling

Abstract: Shallow water acoustic response functions at high frequencies and large bandwidths exhibit spatio- temporal variability that depends greatly on the propagation media's volume and boundary conditions as well as system source-receiver motion For this reason practical acoustic systems invariably must operate without perfect knowledge of the space-time state of the ocean media Considered here is a Gaussian mixture assignment over Doppler and channel bandwidth employed to describe the amplitude and phase of such acoustic response functions over signal duration and bandwidth that can serve in many scenarios to replace recursive least squares and Kalman-like algorithms The mixtue Gaussian model of channel dynamics allows for the accurate and adaptive description of the response function The model is flexible and naturally accommodates varying degrees of observed channel spar- sity Posterior expectations are derived and shown to be soft shrinkage operators over Doppler-channel frequency The model allows for novel and accurate estimates regarding the aggregate acoustic path dilation process that serve to replace conventional phase locked loops This adaptive filtering scheme with aggregate path dilation estimation and compensation is tested on M-ary orthogonal signals at both 1 and 2 bits per symbol during the Unet08 acoustic communication experiments These tests took place in the downward refracting, lossy bottom environment of St Margaret's Bay Nova Scotia off of the R/V Quest Receiver algorithms based on this approach were applied to a single element acoustic time series and empirical bit error rates demonstrate a 4 dB improvement over rank based maximal path combining methods For a single hydrophone at 2 bits per symbol a bit error rate of less than 10-4 is observed at received SNR < -10 dB corresponding to an SNR/bit < 14 dB

A comparison of ISO 9613-2 and advanced calculation methods: Predictions versus experimental results

Abstract: Standardization provides methodologies by which independent investigations of the same situation are able to derive the same conclusions. However, standardization is sometimes also perceived as absolute and accurate, beyond which one should not investigate matters deeper. The responsibility of the accuracy of these methods does not lie with the developers but with the standards organizations. This is not the case with algorithms based on pure scientific research where the full responsibility lies with those who turn it into software applications. The Simulation and prediction of outdoor sound propagation using advanced calculation methods are based on principles of physics with an effort to try to avoid empirical or approximate methods, often found in published outdoor propagation standards.

Preliminary statistical pattern recognition methods in the study of vowels produced by children with and without speech sound disorders

Abstract: The nature of vowel acquisition and vowel error patterns in young children, especially those with speech sound disorders (SSD), is not well understood. Previous studies of vowels produced by children with SSD of unknown origin (SSD-TJNK) have shown that they demonstrate a similar order of acquisition and vowel error patterns as those of children with TDS (typically developing speech), but are slower in developing vowels and produce more vowel errors. Adult participants were 15 women, ages 18 to 35 years. All were monolingual speakers of Western Canadian English with no history of speech delay or disorder. Two groups of children participated (TDS and SSD-UNK), with three children in each of four age groups in each group for a total of 24 children. Vowels of all groups were more successfully classified with two measurements representing the formant movement patterns, than a single point measurement, and a combination of all acoustic measures than a single or a subset of measures.

Hearing loss in classical orchestra musicians

Abstract: Quian et al. performed a noise exposure survey on members of the National Ballet orchestra, using noise dosimeters. A questionnaire was also conducted to gather information on particulars of the surveyed members. Air conducted and bone conducted audiograms were obtained using a clinical audiometer with 5 dB steps. Forty-four of the 52 musicians completed the questionnaire (85%). The majority of respondents reported listening to music through a speaker system (40), while many will also use earbuds with portable listening systems (21). On average the data suggest only a slight to mild mid to high frequency sensory-neural hearing loss in the 4000-6000 Hz region. This is consistent with hearing losses observed in the earlier stages of other forms of noise exposure. for. It is also observed that there are practically no differences between measured and calculated hearing losses at 3000, 4000, and 8000 Hz.

Acoustical Challenges in Long Term Care Facilities

Abstract: Providing satisfactory acoustical environments in healthcare facilities can be ensured by applying recommended minimum design requirements provided in Sound and Vibration Design Guidelines for Hospital and Healthcare Facilities 1 . However, there are still acoustical challenges within long term care facilities that should be addressed, such as: b b &bull Maintaining speech privacy between rooms and corridors while good speech intelligibility through corridors for caregivers to hear calls from residence inside the room is required. I will discuss the aforementioned issues experienced in one Long Term Care (LTC) facilities, including all challenges for improving acoustical separation between a room, holding a person with dementia who was screaming during days and nights, and the public area, TV room / eating room. Proving the steps taken to improve this acoustical separation is discussed including all challenges on how to not affect fire separation of demising walls and door.

The vocal signature of the male northern elephant seal

Abstract: A study that was conducted to examine the vocal signature of the male northern elephant seal is presented. Northern elephant seals breed annually during the winter months at islands and mainland rookeries along the western coast of the United States and Mexico. These animals maintain a highly polygynous breeding system in which adult males establish dominance hierarchies that determine access to estrous females. These complex displays play an important role in settling otherwise costly interactions between competing males, as stereotyped acoustic signals often elicit appropriate behavioral responses from spatially separated individuals without physical contact. Body size parameters for focal individuals were derived through repeated photometric sampling. High- quality, complete calling bouts were digitally recorded and analyzed from sub-adult and adult males to evaluate spectral, temporal, and sound pressure characteristics across both seasons.

Structural analysis of multi-fluid shell systems subjected to an external acoustic pulse

Abstract: An elastic shell filled with and submerged into different fluids is subjected to an external acoustic pulse. It is found that the diversity of the internal reflection and focusing phenomena is such that it leads, for certain combination of the parameters of the fluids, to a very considerable increase of the pressure in the fluid which in some cases can be as high as 110% of the peak incident pressure. A thin elastic circular cylindrical shell filled with and submerged into different fluids is considered. The fluids are assumed to be irrotational, inviscid, and linearly compressible, thus the wave equations are used to model the fluid dynamics. The results show that the tensile stress is affected by the changes of the acoustic properties of the fluids, so much so that it prompts a separate discussion of the matter.

From quantal biomechanics to whole events: Toward a multidimensional model for emergent language

Abstract: The article describes an approach to speech articulation based on a view of vocal tract constrictors as physiological sphincter mechanisms. In this model, a particular constrictor (sphincter) can produce only one kind of constriction, and it does so using the inherent quantal biomechanical properties of sphincters. Within this space, those nonlinearities that facilitate the production of ecologically successful events function as attractors to behavior. These facilitative nonlinearities become the common currency of human ecological space, and coupled with an iterative learning simulator, are the basis for an emergent model of phonology or other patterned behavior. The common currency of this space is communicative advantage, and it is only by offering demonstrative communicative advantage that a behavior may be adopted as exerting an effect.

A comparison of spatial listening in a soundbooth versus an immersive virtual environment

Abstract: A comparative study of spatial listening in a soundbooth versus an immersive virtual environment was conducted to better understand whether the results of soundbooth studies can be generalized to real life. A target sentence and two competing sentences drawn from the sentences recorded by four males for the Coordinate Response Measure were presented concurrently to the listeners from three different locations. All participants completed 8 sessions in each of two conditions. The two conditions differed in terms of the test environment, while each session consisted of 4 blocks of 30 trials. In both test environments, performance was best when listeners were certain about the location of the target and performance declined as location certainty decreased. The largest difference in performance between the two test environment conditions was only 4.6 percentage points.

Emotion co-exists with lexical effects: A case study

Abstract: A study that was conducted to examine how emotion co-exists with lexical effects is presented Monte-Carlo sampling was used to estimate reported p-values First, the dependent variables were modeled individually as a function of Emotion and Speaker to check whether the acoustic characteristics of the recorded speech were indeed modified by the emotional modality Second, the dependent variables were modeled as a function of the predictors lexical Frequency (logged) and Neighborhood Density; the model also accounted for possible interactions of the two lexical predictors with Emotion and Speaker The effect of Neighborhood Density on vowel duration does not interact with lexical Frequency and lexical Frequency is not predictive of vowel duration

Departures from the acoustical parameters in the intonation of South Indian Musical Intervals

Abstract: All evolved music cultures have engaged in the study of musical intervals. Indian musical intervals are melodic. Indian music was not structured acoustically1. But in recent times scholars began to give precise acoustical values for the musical notes. This paper shows the departures from such acoustical parameters and explores the pitch profiles characteristic to the South Indian raga music. It demonstrates the dynamic variability in the intonation of the South Indian musical intervals. A holistic view incorporating hearing and perception are vital to the understanding of culturally sensitive idiomatic pitching of the musical intervals.

A Bayesian framework for geoacoustic inversion of wind-driven ambient noise in shallow water

Abstract: Bayesian inversion is applied to estimate the joint posterior probability density (PPD) of geoacoustic parameters. The PPD is sampled by a reversible-jump Markov chain Monte Carlo (rjMCMC) algorithm, which uses an extended Metropolis-Hasting (MH) criterion that allows trans-D jumps between parameterizations, quantifying the uncertainly due to the lack of knowledge of the model parameterization. Sequential datsets are obtained by discretizing continuous-time recordings of ambient noise. Conventional beamforming was used to estimate the BL at 8 frequencies in the range 550 Hz to 1400 Hz. The BL data at 20 uniformly-spaced grazing angles from 14° to 90° is provided to the sequential Bayesian trans-D Monte Carlo algorithm for estimation of the PPD. The geoacoustic parameters and the depth of acoustic interfaces closely resemble the true profiles.

Evolution of audiometry: Clinical testing of a new tablet audiometer

Abstract: A clinical study was conducted to determine the accuracy of audiometric thresholds measured using tablet audiometry. The results were compared to those obtained with accepted conventional conditioned play audiometry (CPA). The study was conducted on a population of 85 patients with normal or abnormal hearing, aged 3-16, at the Audiology Clinic at the Children's Hospital of Eastern Ontario. All participants completed two audiometric evaluations, one with the tablet audiometer and one with conventional CPA and the order of the test was determined at random. 55 subjects were identified by the conventional CPA test to have normal hearing. Of these, 52 were correctly identified with tablet audiometry, the remaining 3 children scoring slightly outside the parameters defined for normal hearing. For patients who had separate left-right ear assessments, a repeated measures model for the threshold in each ear at each frequency was fitted using linear mixed effects modeling.

The Development of VOT perception in school-aged children

Abstract: Preliminary results of an on-going research project that aims to investigate the development of pre-vocalic stop categorization in children aged 4 to 9 are presented. The study examines whether there exists age-related perceptual differences in school-aged children and investigates whether developmental differences in phonemic categorization vary as a function of stop place of articulation. A total of 48 children aged 4 to 9 participating in the study were self-reported as normal hearing individuals with no known language, speech, or learning delays. Older children demonstrated a sharper transition between phonemic categories as well as a more precise phonemic boundary location. Linguistic boundary solidification in voice onset time (VOT) perception development is similar to the reported production development sequence, with the phonemic categorization pattern being consistent with the suggested sequence of emergence of these articulators.

Prosodic phrasing in NXA?AMXCÍN (Salish) declarative clauses

Abstract: The article reports results from a pilot study of the phonetic correlates of declarative clauses from a narrative in Salish. There is a small but growing body of research describing the properties of prosodic phrasing in Salish languages. While seminal research on prosodic phrasing in English and Japanese is based on targeted elicitations, work on prosody in endangered languages often comes from varying sources. The maximum FO of stressed, phrase-initial and -final vowels of the target phrase, as well as the final vowel of the preceding phrase were measured. If pitch was higher than that of either surrounding vowel, it was considered a pitch peak. Phrase boundaries were correlated with a L% initial and final tone, provided the boundary vowels were not stressed. About 89% of phrases with unstressed boundary vowels were associated with a L%, while 91% of phrases with a stressed boundary vowel showed no rise or fall.

Consonantal duration scaling in accented words

Abstract: A study that was conducted to examine the consonantal duration scaling in accented words is presented. The goal was to capture the durational differences of the different onset consonants of accented words, as the hypothesis is that they will scale differently according to the type of consonant. That is, the onset consonant of an accented word will occupy a larger percentage of the total word duration than it would otherwise occupy in an unaccented position. Fuller constrictions would then likely be more susceptible to holding this pause longer since the closure gestures are not being made in the fricative and liquid consonants. As such, for these other consonants, a different prosodic cue to accented words are likely more relevant- most likely intonation, intensity, or something else in the surrounding environment.

Modelling vowel inherent spectral change in spontaneous speech

Abstract: A study that was conducted to examine the vowel productions in a corpus of spontaneous speech is presented. Several alternate accounts of the nature of VISC in speech production are also briefly discussed. Research on vowel identification has suggested that the onset+offset hypothesis best captures the perceptual cues listeners use. Studies on the production of vowels in careful citation speech have added to this theory with models that include vowel duration and pitch. On the plots, the arrow indicates the average vowel offset and the labeled end represents the average vowel onset. The linear VISC movement shown is gathered from the onset+offset model. It is notable that even though the vowel spaces differ between males and females, the trends in VISC movement remain the same. Most studies on vowel identification make use of a onset+offset theoiy of VISC, with the data here point to the discriminative importance of this hypothesis in spontaneous speech productions.

Alternative source for marine geophysical exploration

Abstract: Fundamental physics issues associated with low frequency sources, particularly when they are operated near the water surface, and why it is so difficult to achieve even a small fraction of the power that an air gun produces with electroacoustic other marine vibroseis technologies, are discussed. The power level achieved at 5m depth is about the same as in the free field at 25Hz, but at lower frequency it is attenuated and at higher frequencies it is boosted by the surface reflection. Studies were conducted to find that low frequency projectors just either produce large displacements or have large radiating areas, or both. The operating frequency of a transducer greatly affects the transducer size and effectiveness. To reduce the operating frequency, transducer is made less stiff. In air backed transducers, such as flex-tensionals, the operating frequency is lowered by reducing the mechanical stiffness of the flexing member.