Search

The hearing sensitivity of a 2-yr-old male harbor porpoise was measured using a standard psycho-acoustic technique under low ambient noise conditions. Auditory sensitivity was measured for narrow-band 1 s sweeps (center frequencies: 0.125-150 kHz). The audiogram was U-shaped; range of best hearing (within 10 dB of maximum sensitivity) was from 13 to ∼140 kHz. Maximum sensitivity (threshold: ∼39 dB re 1 μPa) occurred at 125 kHz at the peak frequency of echolocation pulses produced by harbor porpoises. Reduced sensitivity occurred at 32 and 63 kHz. Sensitivity fell by ∼10 dB per octave below 16 kHz and declined sharply above 125 kHz. Apart from this individual's ca. 10 dB higher sensitivity at 0.250 kHz, ca. 10 dB lower sensitivity at 32 kHz, and ca. 59 dB lower sensitivity at 150 kHz, his audiogram is similar to that of two harbor porpoises tested previously with a similar psycho-acoustic technique.

It is often assumed that listeners detect an increment in the intensity of a pure tone by detecting an increase in the energy falling within the critical band centered on the signal frequency. A noise masker can be used to limit the use of signal energy falling outside of the critical band, but facets of the noise may impact increment detection beyond this intended purpose. The current study evaluated the impact of envelope fluctuation in a noise masker on thresholds for detection of an increment. Thresholds were obtained for detection of an increment in the intensity of a 0.25- or 4-kHz pedestal in quiet and in the presence of noise of varying bandwidth. Results indicate that thresholds for detection of an increment in the intensity of a pure tone increase with increasing bandwidth for an on-frequency noise masker, but are unchanged by an off-frequency noise masker. Neither a model that includes a modulation-filter-bank analysis of envelope modulation nor a model based on discrimination of spectral patterns can account for all aspects of the observed data.

The issue of heavy weight drops in buildings has been an issue for a number of years. Currently, the assessment of the noise and vibration due to heavy weight drops in an existing building can be quite cumbersome. All of the potential floor coverings and the heavy weights need to be shipped to site. Typically, the impact sources used onsite are whatever happens to be present at the site. Overall, this testing regimen is quite arduous, time consuming, and not very repeatable. In recent years, the authors have built a drop tower to repeatedly test fitness flooring performance in a controlled environment. This paper looks at ways to predict the in situ performance of fitness flooring by incorporating force pulse data collected from a drop tower and measured transfer functions. This would be a narrow band analysis that can be converted into one-third octave bands as opposed to the one-third octave band analysis that others have suggested. That analysis used a method similar to the Delta IIC in ASTM E2179.

By directing auditory selective attention, humans can recognize a target sound from multiple environmental sounds. Evidence suggests that the effect of auditory selective attention is observed not only in the frequency domain, but also in the spatial domain. However, the effect of auditory spatial attention is not always observed and has been suggested to depend on the level of auditory processing. Therefore, the task dependency of auditory spatial attention was investigated by measuring the word intelligibility and detection threshold of target 1/12 octave-band noise amongst those with different center frequencies. Target and distractor sounds were presented by a loudspeaker array surrounding listeners. The target sound was presented after listeners' auditory attention was attracted to the specific direction from which the target sounds would be presented. Results showed that the word intelligibility increased when attended but the detection threshold of narrow-band noises did not change significantly. This suggests that auditory processing higher than that required for detecting a specific random noise among spatially distributed similar noises is necessary for the effect auditory spatial attention to appear. It seems interesting to pursue the lowest level, where the effects of auditory spatial attention are observable.By directing auditory selective attention, humans can recognize a target sound from multiple environmental sounds. Evidence suggests that the effect of auditory selective attention is observed not only in the frequency domain, but also in the spatial domain. However, the effect of auditory spatial attention is not always observed and has been suggested to depend on the level of auditory processing. Therefore, the task dependency of auditory spatial attention was investigated by measuring the word intelligibility and detection threshold of target 1/12 octave-band noise amongst those with different center frequencies. Target and distractor sounds were presented by a loudspeaker array surrounding listeners. The target sound was presented after listeners' auditory attention was attracted to the specific direction from which the target sounds would be presented. Results showed that the word intelligibility increased when attended but the detection threshold of narrow-band noises did not change significantly. T...

Toothed whales depend on echolocation for orientation and prey localization, and source parameters of echolocation clicks from free-ranging animals therefore convey valuable information about the acoustic physiology and behavioral ecology of the recorded species. Recordings of wild hourglass (Lagenorhynchus cruciger) and Hector's dolphins (Cephalorhynchus hectori) were made in the Drake Passage (between Tierra del Fuego and the Antarctic Peninsular) and Banks Peninsular (Akaroa Harbour, New Zealand) with a four element hydrophone array. Analysis of source parameters shows that both species produce narrow band high-frequency (NBHF) echolocation clicks. Coastal Hector's dolphins produce clicks with a mean peak frequency of 129 kHz, 3 dB bandwidth of 20 kHz, 57 micros, 10 dB duration, and mean apparent source level (ASL) of 177 dB re 1 microPa(p.-p.). The oceanic hourglass dolphins produce clicks with mean peak frequency of 126 kHz, 3 dB bandwidth of 8 kHz, 116 micros, 10 dB duration, and a mean estimated ASL of 197 dB re 1 microPa(p.-p.). Thus, hourglass dolphins apparently produce clicks of higher source level, which should allow them to detect prey at more than twice the distance compared to Hector's dolphins. The observed source parameter differences within these two NBHF species may be an adaptation to a coastal cluttered environment versus a deep water, pelagic habitat.

The present study was a follow-up to a pilot study in which it was found that a 500-Hz-wide narrow-band noise (NBN) masker produced more masking than a tonal (T) masker for signal frequencies both above and below the masker frequency. The aim of the present study was to determine to what extent these results were influenced by an interaction of the relatively rapid temporal envelope fluctuations of the NBN and the short (10-ms) duration of the signal. In the first experiment, the masking produced by a regular NBN, a low-noise noise (LNN), and a T was compared. The LNN produced less masking than the NBN, and about as much as the T, suggesting that the inherent amplitude fluctuations in the NBN were largely responsible for the greater masking produced by that masker. In the second experiment, the masking produced by a regular NBN was compared with that by a T for a signal duration of 10 or 200 ms. The difference in masking between the two maskers was reduced or eliminated when the signal duration was 200 ms, because the threshold in the presence of the NBN masker decreased more with increasing signal duration. This could reflect a decreased "confusion" between the signal and the inherent fluctuations of the NBN masker.

A "simple" dichotic pitch arises when a single narrow band possesses a different interaural configuration from a surrounding broadband noise whose interaural configuration is uniform and correlated. Such pitches were created by interaurally decorrelating a narrow band (experiment 1) or by giving a narrow band a different interaural time difference from the noise (experiment 2). Using an adaptive forced-choice procedure, listeners adjusted the interaural intensity difference of "pointers" to match their lateralization to that of the dichotic pitches. The primary determinants of lateralization were the interaural configuration of the broadband noise (experiment 1), the center frequency of the narrow band (experiment 1), and its interaural configuration (experiment 2). The ability of two computational models to predict these results was evaluated. A version of the central-spectrum model [J. Raatgever and F. A. Bilsen, J. Acoust. Soc. Am. 80, 429-441 (1986)] incorporating realistic frequency selectivity accounted for the main results of experiment 1 but not experiment 2. A new "reconstruction-comparison" model accounted for the main results of both experiments. To accommodate the variables shown to influence lateralization, this model segregates evidence of the dichotic pitch from the noise, reconstructs the cross-correlogram of the noise, and compares it with the cross-correlogram of the original stimulus.

The masking patterns produced by narrow-band maskers can show distinct irregularities. These experiments attempted to clarify the relative importance of factors contributing to these irregularities. A three-alternative adaptive forced-choice method with feedback was used, to promote use of the optimal detection cues. The masker and signal were either a sinusoid or a band of noise that was 80 Hz wide, giving four possible combinations of masker and signal type. In experiment 1, masking patterns were measured for maskers centered at 1 kHz, for all combinations of masker and signal type (tone or noise). The masking patterns showed irregularities (dips or "shoulders") above the masker frequency, and the irregularities were larger for the sinusoidal than for the noise masker. Experiment 2 was similar to experiment 1, except that low-pass noise was added to mask combination products. For the noise masker, the low-pass noise slightly increased thresholds, and largely eliminated the irregularities in the patterns, but for the tone masker, the irregularities persisted. Experiment 3 used a noise signal with tone and noise maskers centered at 250, 1000, and 4000 Hz. The tone masker produced less masking than the noise masker for masker-signal frequency separations of 150-250 Hz, regardless of masker frequency. Experiment 4 used an additional masker tone to introduce beats similar to those produced by the interaction of the signal and (main) masker, and to mask combination products. This largely eliminated the dips in the masking patterns for both the noise and tone maskers. Experiment 5 used an additional pair of high-frequency tones to introduce beats, with similar results. We conclude that temporal fluctuations (beats) have a strong influence on the masking patterns for sinusoidal maskers, for masker-signal frequency separations up to a few hundred Hz. Beats may also have some influence on the masking patterns for noise maskers. The detection of combination products also plays a role.

The auditory brain‐stem responses (ABRs) to rarefaction and condensation clicks were obtained for 12 normal‐hearing subjects in quiet, and high‐pass masking at 8, 4, 2, 1, and 0.5 kHz. Derived narrow‐band wave V latency differences were analyzed with respect to (1) stimulus polarity, (2) absolute differences irrespective of polarity. The analyses revealed no significant stimulus polarity effects on latency for the derived bands. Absolute latency differences regardless of polarity tended to be greater for those derived bands having lower characteristic frequencies (CFs). However, these differences were smaller than the expected half‐period of the theoretical CF. Further analyses in three additional subjects using repeated runs of the same polarity indicate that this increase in absolute latency difference with lower derived band CF does not reflect a simple half‐period change owing to polarity, but rather to the increase variability in measuring the peak latency of the lower CF derived bands. The variability is consistent with variability of eighth nerve PST histograms behavior observed in animal work [Kiang et al., ‘‘Discharge patterns of single fibers in the cat’s auditory nerve,’’ Research Monograph No. 35 (MIT, Cambridge, MA, 1965)]. Thus claimed polarity effects observed in other ABR work using absolute values may have been affected by this variability. It appears from these current data that half‐period latency shifts of wave V owing to stimulus polarity differences are not observed in derived bands responses initiated from frequency specific regions of the cochlea.

The ability to discriminate spectral changes in narrow‐band sounds on the basis of multiple cues is examined. The task is to discriminate a sound consisting of three equal‐intensity tones with frequencies of 960, 1000, and 1040 Hz from a sound with an increment in the level of the center component. In different conditions, listeners are instructed to discriminate the stimuli according to either ‘‘loudness,’’ ‘‘pitch,’’ or ‘‘roughness’’ cues. Spectral weights are estimated in each condition by the method of COSS analysis [B. G. Berg, J. Acoust. Soc. Am. 86, 1743–1746 (1989)]. Relative to a condition without instructions, systematic changes in the pattern of weight estimates are found for each condition. Weight estimates for the loudness and pitch conditions follow patterns predicted by respective models that base discriminations on overall level and on the envelope‐weighted average of instantaneous frequency [L. L. Feth and H. O’Malley, J. Acoust. Soc. Am. 62, 940–947 (1977)] with off‐frequency listening. Weight estimates for the roughness condition change relative to other conditions, but do not follow the predictions of a model which uses information about the envelope of the temporal waveform. However, when loudness and pitch information are degraded in a post hoc condition, weight estimates are consistent with those predicted by the envelope model. Changes in weight patterns with instructions are interpreted as changes in listeners’ decision strategies among multiple cues. Departures of weight patterns from model predictions suggest that the processes that underlie dimensions of pitch and roughness are not independent of loudness.

Apparently on the basis of room modes being excited in residential dwellings, one concept has been to ignore high quality field measurement recordings (wave files) and use the narrow band (FFT) Leq results of a 10 min sample to create, by superimposing a number of sine waves, a trapezoidal time signal as the source for subjective testing and restricting the bandwidth to only infrasound. Other testing utilizes full spectrum signals but has limitations on having an accurate signal due to the limitation of the speakers. An examination of both methods and the limitations of the results have been examined and will be discussed.

This study concerns two situations that could limit a listener’s sensitivity to interaural temporal delay (ITD). The first is having to detect a ‘‘target’’ ITD in one spectral region when diotic energy is also present in remote and/or adjacent spectral regions. In such a situation, sensitivity to ITD is typically degraded, as compared to when no remote and/or adjacent diotic energy is present. Both the outcome and the paradigm are commonly termed ‘‘spectral interference.’’ In the present study, spectral interference was measured using a broadband (100 Hz to 9 kHz) noise that was diotic, save for a restricted spectral region that was interaurally delayed. The portion of the noise that contained the ITD had a center frequency (CF) of either 300, 1200, 2400, or 4800 Hz and a bandwidth that was 40% of the CF. The second situation of interest was ‘‘spectral uncertainty.’’ Here, the center frequency of the portion of the noise containing the ITD was varied (chosen from one of four possible CFs) on a trial‐by‐trial basis. Consequently, the listener was uncertain about which spectral region could contain the ITD. Data were also collected with one of four narrow bands of noise with the CFs and bandwidths identical to the interaurally delayed spectral regions used in the spectral interference and spectral uncertainty conditions. Within a block of trials, the center frequency of the narrow band of noise was either held constant or was randomly chosen from the set of four CFs. In all cases, threshold ITDs were measured in a single‐interval task. Consistent with previous studies, the largest amounts of spectral interference occurred when the portion of the noise containing the ITD was centered at or above 2400 Hz. The novel finding was that spectral uncertainty produced extremely poor performance, rendering some listeners unable to detect even a 1‐ms ITD presented within high‐frequency regions.

The ability to detect changes in the spectral shape of narrow-band tonal complexes (spectral profiles) is examined. The standard consists of three tones of equal intensity; thresholds for detecting an increment in the level of the central, 1000-Hz tone are estimated. A roving-level procedure is used in order to impose a statistical limit on thresholds that can be obtained by basing discriminations on absolute intensity. Estimated thresholds are consistently below this limit, thus indicating the use of other cues. Generally, thresholds are constant for bandwidths ranging from less than a critical band to greater than several octaves. Spectral weight estimates, however, are highly dependent on bandwidth, providing evidence that the discrimination of narrow-band spectra involves different auditory processes than those used to discriminate wideband spectra. Additional data show that pitch cues are important within a restricted range of intermediate bandwidth, but not for wideband or very narrow-band spectra. A version of the EWAIF model involving off-frequency listening is proposed to account for the results.

Detailed comparison between narrow-band and broadband techniques used in a series of acoustic experiments are presented with emphasis on (a) comparisons of the pressure field versus range predicted, (b) comparisons of the bottom compressional-wave-speed model generated, (c) the variance of the bottom model estimated, and (d) the resolution of the bottom model generated. The range dependence of the bottom medium and the seasonal dependence of the near-surface sediment sound-speed profile are observed and discussed

Hit and false-alarm rates were measured for detection of a 500-Hz tone target in each of ten reproducible samples of 1/3-oct bandwidth noise centered at 500 Hz for both NoS pi and NoSo conditions. The effects on hit rates of the starting phase of the target relative to individual noise samples were investigated with two target phase angles for three subjects. The major results are: (1) performance varies significantly over masker waveforms; (2) for NoS pi conditions, the effect of target-to-marker phase angle on hit rates is not significant for these narrow-band maskers; (3) for NoSo conditions, the target-to-masker phase angle has a large effect; (4) no significant correlation between NoSo performance and NoS pi performance is seen across masker waveforms. These results are generally consistent wuth previously reported results for wideband maskers [R.H. Gilkey, D.E. Robinson, and T.E. Hanna, "Effects of masker waveform and signal-to-masker phase relation on diotic and dichotic masking by reproducible noise," J. Acoust. Soc. Am. 78, 1207-1219 (1985)] with an important exception. Specifically, in the wideband experiment, significant correlation between NoSo and NoS pi performance across noise samples was found. In addition, in the wideband experiment, a small yet statistically significant effect of target-to-masker phase was observed in the NoS pi condition.

Given Kryter’s extensive work on the physiological and subjective effects of noise on the human auditory system it is not surprising that at some point he would turn his attention to the impact of noise on speech intelligibility. In 1962, Kryter published “Methods for the Calculation and Use of the Articulation Index” greatly simplifying the methods originally proposed by Fletcher in the 20s and 30s and French and Steinberg in 1947, ultimately leading to the ANSI 1969 standard. Over the years, the AI and the revised Speech Intelligibility Index (SII) have become the leading metric for predicting speech recognition in noise and in quantifying the audibility provided by amplification. However, the scope of the AI/SII is still limited to communication conditions, which do not include multiple, narrow bands of speech or noise. Similarly, predictions for conditions of auditory-visual speech are only roughly approximated. Data from a variety of studies suggest that rather than ignore these communication conditions they should be studied closely for what they reveal. In this talk, I will show data that expose chinks in the armor of our understanding of speech perception in noise and under conditions where visual speech cues are present.

Information about the front/rear location of a sound source is available in the relationship between the direction of head rotation and the direction of changes in interaural time and level differences (ITD and ILD). Our previous results show that such dynamic cues are highly effective for low-frequency stimuli, but minimally effective for narrowband high-frequency stimuli, in which, respectively, ITD and ILD cues are primarily available. In this study, we assessed the possible benefit for dynamic localization of providing more robust envelope ITD cues in high-frequency stimuli. Listeners judged the front/rear location of anechoic free-field stimuli presented over the central portion of a slow (~0.25 Hz), continual, 90-degree head oscillation. Stimuli were bursts of wideband (0.5–16 kHz), low-frequency (0.5–1 kHz), or high-frequency (6–6.5 kHz) random-phase noise or of raised-sine stimuli with exponent 2, modulation frequency 125 Hz, and bandwidth 6–6.5 kHz. Localization accuracy was high for wideband and...

The aeroacoustic properties of the vocal folds and tract are difficult to measure directly. Here, they were measured using broad- and narrow-band excitation at the mouth during phonation into various acoustic loads, including a non-resonant load provided by an acoustically infinite waveguide with cross section comparable with that of the tract. The tract is treated as a duct terminated by the larynx. Mechanical properties of the walls and terminations were determined using a microphone array [Dickens et al. (2007)]. The vocal fold response was monitored with an electroglottograph and wall motion was measured electromechanically. The impedance spectra show negative resistance bands at frequencies near those of phonation, consistent with regeneration at the folds. The walls give inertances consistent with thicknesses of order 1 cm and compliances consistent with distributed stiffnesses of about 100 kN/m3 [Hanna et al. (2012)]. The duct resonant properties are consistent with losses several times higher than...

There are between 67 and 76 species of odontocetes (toothed whales) and presumably all have biosonar capabilites. There are three fundamental biosonar signal types that can be categorized by types of marine mammals that produce these signals. Whales and dolphins that can emit whistle signals (except for sperm whales) project short broadband clicks containing about 5 to 7 cycles with decaying exponential envelope and Q (center frequency over bandwidth) between 2 and 3. Porpoises do not whistle and produce polycyclic narrow band high frequency biosonar signals with approximately 20 or more cycles with a modified sinusoidal amplitude envelope and Q around 14. Biosonar waveforms of beaked whales (also non-whistling animals) typically have 10-15 cycles with a linear FM component and Q around 4. This presentation will discuss the characteristics of the three different biosonar signal types and suggest some motivation factors involved with the use of the different signals. The type of prey and their habitat will also be included in the discussion. It will be shown that in some cases, the signal type is motivated by anatomical constraints of the odontocete and in other cases, the backscatter characteristics of the prey may be the most important factor.

This paper suggests that narrow band specific information of a room's acoustic behavior can be revealed through the use of musical instruments as impulse sources that would be otherwise buried in response to a broadband impulse excitation or other sequence. An advantage to this is discovering acoustic anomalies before data analysis, as well as subjective judgments to room response and stage acoustics to add to standardized and more advanced technical methods. A violin, snare drum, and double bass are examined in terms of directionality and frequency content with and an example hall is analyzed.

There is a strong need for improved sound source localization software for use by scientist interested in conducting passive acoustic surveys of marine and aquatic habitats where most biological sounds are currently unidentified. Fish sounds are typically low frequency (50–1200 Hz), narrow band, knock trains, short duration grunts, or tones, and can be repeating or irregular. These characteristics together with typically fuzzy signal onset, noisy environmental conditions, and shallow depths often make source localization challenging. We used small 3-D arrays of six hydrophones placed in a fixed-frame square diamond configuration to collect known and unidentified fish sounds for testing from a variety of shallow marine and freshwater habitats. The various methods of sound source localization based on measurements of Differences in Time of Arrival (DTOA) at various hydrophones were investigated, including cross-correlation, first pulse arrival, and phase measurement of tonal components in the fish signal. T...

The various arrays sea test (VAST) was conducted in July 1989 in the northeast Pacific Ocean. A 3000‐m‐long, 200‐element hydrophone array was suspended for an 11‐day period from R/P FLIP, moored about halfway between San Diego and Hawaii (34N, 140W). Simultaneously, 12 neutrally buoyant, freely drifting Swallow floats equipped with both infrasonic hydrophones and three‐component geophones were deployed at various depths 150 km off the California coast (35N, 122W). The statistical properties of deep ocean noise recorded by these sensor systems, both single element and beam level, are examined quantitatively under a variety of conditions, including wind‐dominated (15–18 kt), distant air gun operations, and marine mammal vocalizations. Non‐parametric statistical tests applied to narrow band complex time series of the data include the Wald–Wolfowitz run test for mutual independence of the data samples, the Kolmogorov–Smirnov two‐sample test for stationarity, and the Kolmogorov–Smirnov one‐sample test for Gaus...

Dall’s porpoise (Phocoenoides dalli) echolocation clicks have not been widely recorded. Concurrent with visual observations, acoustic recordings of free‐ranging Dall’s porpoise were made offshore of southern California using a towed hydrophone array with two elements of 250 kHz bandwidth. We examined 6035 clicks from 12 sessions totaling more than two hours over the course of seven days. The Dall’s porpoise echolocations recorded were short (48–804 μs), narrow band (2–10 kHz [−3dB]) clicks with most peak frequencies between 117 and 141 kHz, but some as high as 198 kHz. Many clicks contained a multipulse temporal structure, resulting in stereotyped spectral peaks and notches. Two distinctive click types with different spectral banding patterns and peak frequencies (122.8 and 135.8 kHz) were observed. Spectral banding patterns have been used as a species identifier for Risso’s dolphins and Pacific white‐sided dolphins. These two dolphins and Dall’s porpoise have similar head morphologies, which may play a role in producing clicks with spectral peaks and notches. This study shows that Dall’s porpoise produce multiple click types, which may provide a tool for population classification, and that their clicks contain spectral banding patterns, which may provide insight into the mechanism by which such clicks are produced.

The purpose of this study is to investigate if comodulation masking release experiments with hearing impaired (HI) individuals can be used as a reliable method to detect dead regions in the cochlea, if any. The experiment involves detection of a pure tone in a narrow band of noise (the on‐signal band) in presence of four other narrow bands of noise that fall outside the auditory bandwidth at the center frequency of the on‐signal band. These flanking bands are either all co‐modulated so as to have the same amplitude envelope as the on‐signal band or are randomly modulated and are presented monaurally to an impaired ear. Each band is level adjusted to be audible to the HI ear. Data are being collected for both these cases from normal‐hearing listeners and HI listeners with no dead regions. Every HI individual also undergoes psychophysical tuning curve and threshold‐equalizing noise tests which are currently used tests to detect cochlear dead regions. Each HI individual also takes speech perception tests wit...

In a recent paper by Moore, Alcantara, and Dau [J. Acoust. Soc. Am. 104, 1023–1038 (1998)] masking patterns for sinusoidal and narrow‐band noise were presented. Data were collected for all four masker‐signal combinations. The same subjects participated in the different experimental conditions and the same experimental procedure was used in each case. Despite the influence of combination products, a strong influence of temporal fluctuations on the masking patterns for sinusoidal maskers was found for masker‐signal frequency separations up to a few hundred Hz. The present study shows model predictions obtained with the modulation‐filterbank model [Dau et al., J. Acoust. Soc. Am. 102, 2892–2905 (1997)]. This model is able to use envelope fluctuations, for example, beats, as a detection cue, but it does not contain a nonlinear peripheral filter stage. Thus the model predictions offer the possibility to separate between effects caused by temporal effects like beating and those caused by combination products. As in the experimental data of Moore et al., less masking was found for the tone‐on‐tone masking condition than for the tone‐in‐noise masking condition, for tone‐masker frequency separations lower than 250 Hz. The asymmetry of masking between the signal‐tone masker‐noise condition and the signal‐noise masker‐tone condition can also be accounted for quantitatively by the model.

The source properties of toothed whale echolocation clicks are essential for evaluating the performance of biosonar systems. However, such data only exist for a small number of free ranging toothed whale species. Here we present a quantitative analysis of the first recordings of echolocation sounds from hour glass dolphins (Lagenorhynchus cruciger) made in the Drake Strait. Recordings were made with a vertical array of four broad‐band, calibrated hydrophones that allows for acoustic localization and source parameter derivation. Only apparent on‐axis clicks were used for analysis. Hour glass dolphins produce clicks with a mean peak and centroid frequency around 128kHz, a narrow, mean RMS bandwidth of 10kHz with no energy below 100kHz and a mean signal duration of 161μs. Echolocation clicks of hour glass dolphins are thus of the high frequency narrow band‐width type (NBHF type) found in porpoises, a few delphinids of the Cephalorhynchus genus and the pygmy sperm whale Kogia breviceps, but are strikingly dif...

Speech quality of telecommunications systems is usually evaluated thanks to auditory tests, which must be carried out in accordance with ITU‐T Recommendations. In these tests, subjects are asked to assess the quality of speech sample by giving a score on a five‐level scale. The averaging of subjects' scores yields the Mean Opinion Score (MOS) which represents the speech quality for a given condition. However, MOS values can be strongly influenced by many factors. In this paper, we focus on a specific bias: the distribution of the impairments in the stimuli set. In effect the rating of a narrow‐band stimulus (300‐3400 Hz) may have a lower quality score in a mixed‐band set (mixed of narrow‐band and wideband (50‐7000 Hz) conditions) than in a purely narrow band set. This effect is detailed through results of auditory tests recently carried out. Consequently, the validity of MOS value is theoretically limited within a test which prevents MOS comparison between different tests. Finally, two solutions are propo...

Noise stimuli were used to investigate weighting patterns for detecting dynamic frequency changes. For one stimulus set, a tonal glide was modulated by a narrow band of noise (gliding narrow‐band noise, GNBN, condition). For the second set, the glide was preceded and followed by steady‐state tones before modulation (SS condition). The task was to indicate whether the stimulus had a frequency glide or not. Weighting patterns (as a function of time) were derived for each subject based on a linear classification model. First, the deviation of the instantaneous carrier for each stimulus was calculated and then weighted by the squared envelope of the stimulus. This was followed by the linear classifications of the squared‐envelope‐weighted deviation of instantaneous carrier (SEWDOIC) according to the subjects responses [A. Ahumada, J. Vis., 2, 121–131 (2002)]. The weighting patterns were consistent across subjects although individual differences in pattern details were observed. The comparison between the patterns from the GNBN and SS conditions indicates the use of dynamic information rather than only comparing the beginning and ending frequencies, consistent with Lyzenga et al. [J. Acoust. Soc. Am. 116, 491–501 (2004)]. Similar weighting patterns were also estimated when the center frequency of the stimuli was random.

The spectral ‘‘center‐of‐gravity’’ (COG) hypothesis proposed by Chistovich and colleagues for static vowels suggested spectral auditory integration of two or more formant peaks within 3.5 barks. More recently, studies have examined dynamic COG effects by modifying the amplitude of steady‐state formants over time in order to create ‘‘virtual’’ diphthongs. The present study looks at ‘‘virtual frequency change’’ cues to the alveolar‐velar place distinction in stops [an extension of Fox and Smith, J. Acoust. Soc. Am. 117, 2423 (2005)]. First, [d≳‐g≳] and [t≳‐k≳] series were synthesized with a 50‐ms F3 transition (with a voiced or voiceless source). Next, matched stimulus series were created with virtual F3 transitions by changing the amplitudes of either a pair of sine waves (in [d≳]‐[g≳]) or a pair of narrow band noise signals (in [t≳]‐[k≳]) whose frequencies did not change. Another matched [t≳]‐[k≳] series was produced with both actual and virtual stops bursts (with no F3 transitions). Identification of the...

To elucidate the roles of envelope and carrier cues in detection of a 500‐Hz tone in 50‐Hz bandwidth noise under N0Sπ conditions, hit‐ and false‐alarm rates were collected for 4 sets of reproducible stimuli. The envelopes and carriers from two sets of stimuli (each with 25 signal‐plus‐noise and 25 noise‐alone waveforms) were extracted and recombined to form 2 new sets (each having the carriers from one of the original sets and the envelopes from the other). Noises were preselected to produce minimal spectral splatter in the recombination processes. Preliminary results for signal‐plus‐noise trials suggest that carrier cues (e.g., temporal fine structure) are dominant over envelope cues (i.e., response patterns were more similar between sets with the same carriers but different envelopes, than between sets with the same envelopes but different carriers); however, carrier cues cannot completely explain the data. Results for noise‐alone trials suggest that both envelope cues and carrier cues contribute to listener responses. Further analyses will explore potential interactions between monaural and binaural stimulus properties, as well as interactions between interaural envelope cues and carrier cues. [Supported by NIH DC007798(SAD), DC001641(LHC), and DC00100(HSC).]

This investigation analyzes measurements of the reflected sound field from various reflector panel arrays. A previous study by the third author [R. R. Torres and M. Vorlaender, ‘‘Scale‐model MLS‐measurements of scattering from overhead panel arrays,’’ Acta Acustica (in press)] measured impulse responses using the maximum length sequence method at various receiver positions from six scale‐model panel arrays of different sizes and densities. This study included an octave band analysis of reflected magnitude as a function of receiver position. The current authors have extended the research by conducting a narrow‐band analysis of magnitude and phase of the reflected sound fields for three of the panel arrays [J. Rathsam, L. M. Wang, and R. R. Torres, J. Acoust. Soc. Am. 117, 2499(A) (2005)]. This study revealed a close relationship between phase and magnitude. The current work continues the analysis of reflected phase and magnitude for the three additional reflector panel arrays. The newer arrays are more com...

Loudness matches between stimuli that differ in frequency are difficult, presumably due to the strong pitch differences which make it challenging to focus on loudness exclusively. It is hypothesized that the use of stimuli with a less salient pitch percept (but that still permit relatively frequency‐specific excitation of the cochlea) would allow for more accurate across‐frequency loudness matches. This would permit more stable comparisons of loudness growth between frequency regions (corresponding to normal and elevated thresholds) as is often the need in studies involving hearing‐impaired subjects. In this study, normal‐hearing listeners were evaluated in loudness matching tasks with two types of stimuli: pure tones and narrow bands of noise (with corresponding center frequencies). Listeners matched the loudness of a stimulus in one frequency region with a stimulus in another frequency region. The stimuli were either two pure tones, two narrow‐band noises, or a pure tone and a narrow‐band noise. The results will be evaluated based on predictions from an excitation pattern model and transitivity to determine if more accurate predictions are possible using narrow‐band noise stimuli. [Work supported by the University of Minnesota.]

This invention includes an Acousto-optical tuning device by employing the TeO2 in share mode, non-collinear and in high RF frequency (or large Bragg's deflection region), it is able to achieve the telecomm worthy tunable filter that provides high tuning speed, no-moving parts, very narrow filter line-width (0.2 nm for instance) and with miniature size practical for a optical telecom active module. This tuning characteristic is achieved by taking the particular advantages that the angle of incidence (AOI) and angle of diffraction (AOD) are divergent in a frequency range just suitable for a shorter wavelength application required by modern telecommunication systems while the acoustic wave decay in this frequency range can be properly managed to achieve the design goals. Furthermore, by generating a diffracted beam with larger AOD, implementation of filtering for optical propagations of different wavelengths can be conveniently selected. Other than the TeO2 crystals, the AOTF can also be formed with AsS3 and GeAsSe crystals based on the same acoustic-optical interaction processes at a higher RF frequency range as discussed in the above descriptions.

The present study evaluates the relationship between loudness matches and simple reaction times (RT) in six normal‐hearing subjects. Loudness matches between a narrow‐band noise (125 Hz wide) and a broadband noise (1500 Hz) centered at 1000 Hz were made at levels from near threshold to near 100 dB SPL. For the same noises and level range, RT were also measured. In agreement with previous loudness‐matching studies, as SPL increased the level difference between the noises needed to maintain equal loudness first increased, to around 10 dB at moderate SPLs, and then decreased. Except for one listener, the RT data show the same pattern. The level difference needed to maintain RT the same to the two noises first increased and then decreased. These results show that RT is closely related to loudness, but not to sensation level. If RT depended on sensation level, the level difference between the two noises needed to achieve equal RT would not change with SPL because the difference in sensation level between two sounds is constant as a function of SPL. Overall, the present data provide strong support for the contention that simple RT depends strongly on loudness. [Supported by NIH/NIDCD Grant No. R01 DC 02241.]

In order to investigate the effect of sound incidence angle on loudness across a larger set of parameters than have been used in most previous studies, a listening experiment was carried out using a loudspeaker setup in an anechoic chamber. Eight subjects, whose absolute hearing thresholds and head‐related transfer functions (HRTFs) were measured, participated in a total of 22 sessions each. On each trial their task was to judge which of two narrow‐band noises sounded louder. These judgments were used in an adaptive procedure to find loudness matches between a frontal reference location and seven other sources, positioned both in the horizontal and median planes. Sound incidence angle, center frequency, and overall SPL were varied in the procedure. The results show that loudness is not constant over sound incidence angles, with matches varying over a range of 10 dB, and showing considerable frequency dependency. The pattern of results also varies substantially between subjects, but can be accounted for by...

In order to test monaural and binaural processing components of binaural detection, diotic and monaural detection were studied using reproducible noises. Correlations across masker bandwidth and interaural configuration were computed. Preliminary results reveal high correlations for probability of detection across N0S0 and NmSm configurations for both wideband and narrow‐band maskers, suggesting that the binaural system does not influence results across reproducible noises for diotic detection. As reported previously [Evilsizer et al., J. Acoust. Soc. Am. 111, 336–345 (2001)], significant differences in correlations between wideband and narrow‐band results were observed for the diotic case, suggesting cross‐frequency interactions. Preliminary findings indicate that the same result holds for the monaural case. These data will be used to test models that incorporate monaural processing based on the phase‐opponency model [Carney et al., Acta Acustica United with Acustica 88, 334–347 (2002)], and binaural pro...

In the stratified random medium considered here, the spatially fluctuating wave speed and refractivity vary only in the range direction, which is normal to the planes of stratification. To treat backscattering in such a medium, it is convenient to split the field into forward‐ and backward‐propagating components, and derive coupled equations that govern the range evolution of the statistical moments of these components. The derivation entails an ensemble‐averaging operation, which is here predicated on the key assumption that the two components, which are incident from opposite directions on a differential range slab, are statistically independent of the refractivity fluctuations within that slab. The assumption is justifiable when the fluctuations exhibit a sufficiently rapid time dependence. This procedure is used to derive equations for the bichromatic coherence of the two field components, which are then solved to determine the evolution of the power flux associated with the propagation of a signal due to a planar narrow‐band pulse incident on a semi‐infinite scattering slab. It is shown that the pulse signal is attenuated exponentially in the range direction, on the so‐called ‘‘localization’’ scale, while the scattered power forms echo pulses which are gradually attenuated until all the incident pulse energy is reflected back out of the slab.

Time reversal in random media can result in refocusing on the source with a resolution that is better than the diffraction limit. This is because of the multipathing that occurs in the random medium, making the time reversal array appear to be wider than it is. What is not clear, however, is why this phenomenon is statistically stable. That is, why the refocused spot size does not fluctuate in the random medium. The reason for this statistical stability is that it is a time domain phenomenon and does not occur with narrow band signals. I will explain this effect analytically and will also show how it can be used for imaging in random media.

The intelligibility of notch‐filtered sentences may improve by 10% or more when a large spectral gap is filled with noise rather than silence [Warren et al., P&P 59, 275–283 (1997)]. The present study shows that a comparable ‘‘spectral completion’’ (SC) effect occurs when a narrow band of speech is surrounded by complementary noise. In Experiment 1, intelligibility of 2/3‐octave everyday sentences, centered at 1.5 kHz [2000‐order FIR filtering, slopes greater than 1000 dB/octave], was 65% without noise, 72% with complementary low‐pass or high‐pass noise, and 76% with both noise bands present (noise spectrum level at 20 dB relative to the speech). Experiment 2 found a comparable SC effect with partially masked speech: The 2/3‐octave sentences were about 40% intelligible when masked by narrow‐band noise having the same spectral limits (relative spectrum level again 20 dB), but were 50% intelligible when the noise was broadband. Hence, out‐of‐band noise enables listeners to make better use of information within narrow‐band speech, perhaps by masking an otherwise misleading absence of energy in normally coactive regions of the spectrum. Under partial masking conditions, this SC effect can produce an apparent release from masking in the absence of comodulation. [Work supported by NIH.]

By the end of the 1940s naval experience in World War II and progress in undersea physics and oceanogoraphy had brought about recognition of low frequency as a means of passive detection. The discovery of the deep sound channel was published in 1948, defining its role in long range propagation. In January 1950, the National Research Council’s Committee on Undersea Warfare recommended a study be made of these emerging concepts. What came to be known as the Hartwell Committee was formed. The Committee’s report included specific recommendation of real‐time narrow band analysis of radiated sounds and exploitation of low‐frequency signals. In October, the Chief of Naval Research was directed ‘‘to proceed with arrangements with Bell Laboratories to institute a program of research and development in the field of low frequency.’’ This directive, shared with naval laboratories and affiliated universities, initiated an intensive program of sea experiments that continued over four decades. The measurements were paralleled by computer codes developed to analyze and deconstruct data, and eventually model propagation and arrival structure in realistic, range‐varying ways. The sequence of the discovery and validation of features that define the nature of undersea sound as now understood are outlined and discussed.

A WKB approximation for the range dependent modal group slownesses and dispersions in a range dependent waveguide is derived. The resulting slownesses can be used to synthesize time series in a narrow band or over broader bandwidths by adding narrow band results together coherently. The slownesses themselves are also highly useful intermediate products which help give insight into the speeds with which the modes of a range dependent waveguide propagate. The range dependent modal group slownesses are compared to their adiabatic approximations and conclusions are drawn regarding the relative accuracy of these two approximations for time series synthesis.

This study explores the factors that influence the masking of narrow‐band signals in reverberant sound fields. Rectangular rooms were simulated using the image method with a rigid sphere modeling the head. The signals were third‐octave noise bands presented in a continuous broadband noise masker. For a given room simulation, the signals and masker were convolved with impulse responses obtained in the room at multiple azimuth angles and distances, for subsequent presentation via headphones. Adaptive forced‐choice procedures were used with the resulting signals to find monaural and binaural thresholds in listeners with normal hearing. The benefit of separating the masker and signal spatially in the simulation was reduced by reverberation, as expected, although the effects were complex. Among the predictions confirmed by these experiments is that reverberation actually improves thresholds when the masker is closer to the listener than the signal, apparently due to both an increase in monaural signal power a decrease in interaural signal correlation. With masked thresholds obtained over the frequency range of the articulation index, the results of these experiments will allow quantitative predictions of speech intelligibility as a function of room charcteristics and speech and noise source locations. [Work supported by NIH.]

The precedence effect has been shown to be strong for stimuli with onset transients, e.g., clicks. Zurek [J. Acoust. Soc. Am. 67, 952–964 (1980)] showed that a continuous, broadband noise whose fine structure leads that of a simulated reflection dominates lateralization even when onset gating cues have been removed. We report on experiments similar to those of Zurek, but using narrow‐band noises of varying bandwidths and center frequencies. In a 2IFC headphone task, the leading and lagging stimuli are placed on either side of midline (±300‐μs ITD) in the first interval, and the signals to the two ears are switched in the second interval. Subjects report whether the stimulus in the second interval is to the right or left of the first. Results are consistent with a precedence effect for bandwidths as narrow as a critical band. For narrower bandwidths, onset dominance is still observed, but with oscillations in lateralization consistent with the interaural differences expected for narrow‐band stimuli, given ...

The detection of amplitude modulation of a sinusoidal (target) carrier can be impaired by the presence of a modulated sound (interferer) with a different carrier frequency, referred to as modulation detection interference (MDI). Typically, sinusoidally amplitude‐modulated tones remote in frequency are used as the interferer, and the target modulation threshold is increased if the interferer‐ and target‐modulation rates are similar. In the present study, narrow‐band noise was used as the interferer. Detection thresholds for modulations of 5, 10, and 50 Hz imposed on tone carriers were measured as a function of noise bandwidth (0–200 Hz), the spectral distance between the carrier and interferer (0.5–4 octaves), the carrier level (40–80 dB SPL), and the carrier‐interferer level difference (0–30 dB). In all conditions, the intrinsic fluctuations of the noise interfered with modulation detection, with maximum interference at noise bandwidths about twice the modulation frequency. This bandwidth dependence corre...

Using a 21FC paradigm, psychometric functions for the detectability of a tonal signal added to a 40‐Hz‐wide band of Gaussian noise was measured using signal and center frequencies of 600, 1800, and 5400 Hz. In a second condition, the noise‐alone and the tone‐plus‐noise stimuli were scaled so as to yield identical energy values across the two intervals of each trial. Under the assumption of optimal combination of independent cues, a comparison of these conditions allows an estimate of the relative contribution of energy‐based and non‐energy‐based cues for the detection of a tone added to noise. In a third condition, the envelopes of the scaled noise‐alone and tone‐plus‐noise stimuli were extracted, and used to modulate tones of either 600, 1800, or 5400 Hz. The ability to discriminate between envelope patterns was measured, and that measure used to estimate each observer's reliance on envelope‐based cues. Large individual differences were obtained. For one observer, the detection of a tone added to Gaussian noise was determined to rely equally on envelope‐based and energy‐based cues. A second observer was found to depend almost wholly on changes in energy. None of the observers appeared to rely on changes in fine structure that occur when a tone is added to noise, and consistent with that observation, no differential effect of frequency region was obtained. [Work supported by the National Institutes of Health.]

Detection thresholds were measured as a function of the temporal position of a 5‐ms sinusoidal signal presented within one of two 500‐ms observation intervals. Both intervals contained a 400‐Hz‐wide noise masker, logarithmically centered at 2500 Hz, that was gated on at 150 ms and off at 350 ms within each interval. The signal and masker were both gated with 2‐ms linear ramps. The level of the signal was tracked adaptively in blocks of forty 2I 2AFC trials to estimate a 70.7% correct detection threshold. Thresholds were measured at 43 temporal signal positions within the observation interval, with detailed sampling around the onset and offset of the masker to assess temporal edge effects. Experiments to date with four listeners have focused on signals presented at 1900 or 2100 Hz. The temporal microstructure of the detection functions has characteristically revealed a complex pattern of under‐ and over‐shoot (i.e., enhanced and diminished detection, respectively). These effects, which have been observed a...

Several approaches to passive broadband matched‐field processing have been proposed. In most cases, an incoherent average over a narrow band is quite successful in driving down unwanted sidelobes. The ability of coherent broadband techniques to out perform this simple approach may require additional phase information about the source that, in general, may not be available. In this paper, a fully coherent method that can out perform an incoherent average is derived, although its practical application is problematic. In other words, the simple incoherent average may be the most efficient approach. These ideas are demonstrated using data from the SWellEx‐96 experiment. Given the environmental conditions under which these data were taken, the new coherent approach has some physical advantages over other broadband methods since it more efficiently uses variations of range dependence exhibited by the different frequency components of the signal. The net result is that the range, depth, and azimuth ambiguities have comparable spatial extent.

Sentences can retain intelligibility when heard through narrow spectral slits even though the fluctuating amplitude envelope within the narrow frequency bands contains little spectral information [K. R. Riener et al., J. Acoust. Soc. Am. 91, 2339(A) (1992); R. M. Warren et al., Percept. Psychophys. 57, 175–182 (1995)]. Other research has demonstrated that intelligibility is retained when broadband speech is divided into contiguous spectral regions and used to modulate the amplitude of corresponding regions of noise [R. V. Shannon et al., Science 270, 303–304 (1995)]. The current report combines aspects of these previous studies and provides information concerning the use of temporal pattern information in various regions of the spectrum: Two noncontiguous narrow bands of speech having various center frequencies were used separately to modulate the amplitude of noises, which were then refiltered to corresponding center frequencies and bandwidths. Despite the absence of spectral information within each of t...

This study was designed to extent the range of auditory stimuli used to study magnitude estimation scaling of loudness. The five stimuli chosen were: a 1000‐Hz pure tone; narrow band noise (700–1300 Hz bandwidth); broadband noise (100–10 000 Hz bandwidth); rock music [Led Zeppelin, Atlantic Recording Corp., CD Recording No. 19127‐2 (1969)]; and babble speech [D. Kalikow, K. Stevens, and L. Elliott, J. Acoust. Soc. Am. 61, 1337–1351 (1977)]. Subjects were 30 normal young adult women. During the auditory magnitude estimation task for each of the five stimuli, a subject was instructed to assign numers to that simulus which was presented in a randomly ordered series of nine sensation levels (10, 20, 30, 40, 50, 60, 70, 80, and 90 dB SL). An analysis of variance repeated measured design showed no difference in the numerical responses of the subjects for the five different stimuli. Results suggest the presence of an underlying stabilizing factor (internal scaling mechanism) that allows adult subjects to consistently scale loudness irrespective of the type of auditory stimulus presented [J. Zwislocki and D. Goodman, Percept. Psychophys. 28, 28–38 (1980)].

The ‘‘toughening’’ effect in chinchillas exposed to interrupted impact noises was explored as the center frequency (CF) of a narrow‐band (400 Hz), 115‐dB peak SPL impact exposure was varied. CFs were: 0.5, 1.0, 2.0, 4.0, and 8.0 kHz; impacts were presented 1/s, 24 h/day, for 5 days or 6 h/day, for 20 days. Each pair of exposures had equal energy. Results: Initial threshold shifts (TS) for both the 5‐ and 20‐day groups were similar; 40 to 50 dB at the most affected audiometric test frequency. ‘‘Toughening’’ in the interrupted exposures amounted to at most 30 dB. The 1‐kHz impact was most effective in producing ‘‘toughening’’ at all audiometric test frequencies that showed a TS. The maximum effect was seen at 4 kHz; two octaves above the impact CF. For impact CFs ≳1.0 kHz the ‘‘toughening’’ effect, when it was measured, was small (