Firefighters' hearing: a comparison with population databases from the International Standards Organization.

We investigated firefighters' hearing relative to general population data to adjust for age-expected hearing loss. For five groups of male firefighters with increasing mean ages, we compared their hearing thresholds at the 50th and 90th percentiles with normative and age- and sex-matched hearing data from the International Standards Organization (databases A and B). At the 50th percentile, from a mean age of 28 to a mean age of 53 years, relative to databases A and B, the firefighters lost an excess of 19 to 23 dB, 20 to 23 dB, and 16 to 19 dB at 3000, 4000, and 6000 Hz, respectively. At the 90th percentile, from a mean age of 28 to a mean age of 53 years, relative to databases A and B, the firefighters lost an excess of 12 to 20 dB, 38 to 44 dB, 41 to 45 dB, and 22 to 28 dB at 2000, 3000, 4000, and 6000 Hz, respectively. The results are consistent with accelerated hearing loss in excess of age-expected loss among the firefighters, especially at or above the 90th percentile.

Spatial release from informational masking in speech recognition.

Three experiments were conducted to determine the extent to which perceived separation of speech and interference improves speech recognition in the free field. Target speech stimuli were 320 grammatically correct but nonmeaningful sentences spoken by a female talker. In the first experiment the interference was a recording of either one or two female talkers reciting a continuous stream of similar nonmeaningful sentences. The target talker was always presented from a loudspeaker directly in front (0 degrees). The interference was either presented from the front loudspeaker (the F-F condition) or from both a right loudspeaker (60 degrees) and the front loudspeaker, with the right leading the front by 4 ms (the F-RF condition). Due to the precedence effect, the interference in the F-RF condition was perceived to be well to the right, while the target talker was heard from the front. For both the single-talker and two-talker interference, there was a sizable improvement in speech recognition in the F-RF condition compared with the F-F condition. However, a second experiment showed that there was no F-RF advantage when the interference was noise modulated by the single- or multi-channel envelope of the two-talker masker. Results of the third experiment indicated that the advantage of perceived separation is not limited to conditions where the interfering speech is understandable.

The role of perceived spatial separation in the unmasking of speech.

Spatial separation of speech and noise in an anechoic space creates a release from masking that often improves speech intelligibility. However, the masking release is severely reduced in reverberant spaces. This study investigated whether the distinct and separate localization of speech and interference provides any perceptual advantage that, due to the precedence effect, is not degraded by reflections. Listeners' identification of nonsense sentences spoken by a female talker was measured in the presence of either speech-spectrum noise or other sentences spoken by a second female talker. Target and interference stimuli were presented in an anechoic chamber from loudspeakers directly in front and 60 degrees to the right in single-source and precedence-effect (lead-lag) conditions. For speech-spectrum noise, the spatial separation advantage for speech recognition (8 dB) was predictable from articulation index computations based on measured release from masking for narrow-band stimuli. The spatial separation advantage was only 1 dB in the lead-lag condition, despite the fact that a large perceptual separation was produced by the precedence effect. For the female talker interference, a much larger advantage occurred, apparently because informational masking was reduced by differences in perceived locations of target and interference.

The influence of broadband noise on the precedence effect.

This study examined the influence of background noise on the localization of lead-lag noise-burst pairs and on echo threshold for the same stimuli. Experiments were conducted in an anechoic chamber with a leading stimulus delivered from a loudspeaker at 45 degrees to the right of center and a lagging stimulus from 45 degrees left. Lead and lag stimuli were 4-ms bursts of Gaussian noise. In experiment 1, with lead and lag at equal levels and the lag delay fixed at 2 ms, the perceived location of the image produced by the lead-lag stimulus was estimated from subjects' left/right judgments relative to bursts from a "comparitor" loudspeaker whose position could be adjusted. In experiment 2, the dominance of the leading burst on perceived location was measured by determining the increase in level of the lagging burst necessary to produce an image estimated to be centered at 0 degree azimuth. Experiment 3 was concerned with echo threshold. Subjects reported whether or not they heard a sound in the vicinity of the lagging loudspeaker as the lag-burst delay was varied. In all three experiments, data were obtained for four to five stimulus levels in quiet and for three levels of background white noise from a loudspeaker at 180 degrees. The results revealed a substantial weakening effect of background noise on the precedence effect in experiments 2 and 3, and a nonsignificant effect in experiment 1.

Sudden changes in spectrum of an echo cause a breakdown of the precedence effect.

The effect of changing the frequency components of an echo relative to the sound source was examined in a two-choice discrimination task. Subjects sat in an anechoic chamber and discriminated the direction of the lag noise burst within a lead-lag pair presented over loudspeakers. The leading noise burst was broadband, and the lagging burst was either high- or low-pass filtered. On some conditions, this test burst pair was preceded by a conditioning train of burst pairs, which also had a broadband lead and either a high- or low-frequency lag. When the frequency content of the echo was held constant across the conditioning train and test burst pair, echo suppression that was built up during the repeating train was maintained for the test burst pair, shown by the subjects' poor performance in detecting the location of the lagging burst. By comparison, subjects had little difficulty in localizing the lagging burst when the frequency content of the echo changed between the conditioning train and the test burst, indicating that any buildup of suppression during the train was broken when the lagging burst's spectrum shifted. The data are consistent with an interpretation in which echo suppression is temporarily broken when listeners' built-up expectations about room acoustics are violated.

Intensity discrimination for precedence effect stimuli.

When two identical stimuli are presented from two loudspeakers with a brief delay between them, a single image is heard near the source of the leading sound. The delayed sound or echo appears to be suppressed whereas the preceding sound determines perceived location, hence the name, the precedence effect. This study investigated normal-hearing listeners' sensitivity to changes in the intensity of the lagging sound. Pairs of 2-ms white noise bursts, with a 2-ms delay between the onsets of lead and lag, were presented from two loudspeakers 45 degrees left and right of midline in an anechoic chamber. A 2AFC procedure was used to test discrimination of intensity changes in the lead, lag, and both sounds together. The untreated results showed discrimination to be poorest for changes in the lag stimulus. However, when the intensity differences were transformed into predictions of equivalent monaural level based on KEMAR measurements and binaural loudness summation, discrimination for the lag was equal to the other two conditions. A follow-up experiment found that listeners were highly sensitive to the presence of the lag, more sensitive than would be predicted from loudness changes. It is concluded that the precedence effect does not consist of a general suppression or attenuation of the lagging sound, but rather that suppression may be limited to directionality cues.

Onset dominance in lateralization.

Saberi and Perrott [Acustica 81, 272-275 (1995)] found that the in-head lateralization of a relatively long-duration pulse train could be controlled by the interaural delay of the single pulse pair that occurs at onset. The present study examined this further, using an acoustic pointer measure of lateralization, with stimulus manipulations designed to determine conditions under which lateralization was consistent with the interaural onset delay. The present stimuli were wideband pulse trains, noise-burst trains, and inharmonic complexes, 250 ms in duration, chosen for the ease with which interaural delays and correlations of select temporal segments of the stimulus could be manipulated. The stimulus factors studied were the periodicity of the ongoing part of the signal as well as the multiplicity and ambiguity of interaural delays. The results, in general, showed that the interaural onset delay controlled lateralization when the steady state binaural cues were relatively weak, either because the spectral components were only sparsely distributed across frequency or because the interaural time delays were ambiguous. Onset dominance can be disrupted by sudden stimulus changes within the train, and several examples of such changes are described. Individual subjects showed strong left-right asymmetries in onset effectiveness. The results have implications for understanding how onset and ongoing interaural delay cues contribute to the location estimates formed by the binaural auditory system.

Consonant confusions in amplitude-expanded speech.

The perceptual consequences of expanding the amplitude variations in speech were studied under conditions in which spectral information was obscured by signal correlated noise that had an envelope correlated with the speech envelope, but had a flat amplitude spectrum. The noise samples, created individually from 22 vowel-consonant-vowel nonsense words, were used as maskers of those words, with signal-to-noise ratios ranging from -15 to 0 dB. Amplitude expansion was by a factor of 3.0 in terms of decibels. In the first experiment, presentation level for speech peaks was 80 dB SPL. Consonant recognition performance for expanded speech by 50 listeners with normal hearing was as much as 30 percentage points poorer than for unexpanded speech and the types of errors were dramatically different, especially in the midrange of S-N ratios. In a second experiment presentation level was varied to determine whether reductions in consonant levels produced by expansion were responsible for the differences between conditions. Recognition performance for unexpanded speech at 40 dB SPL was nearly equivalent to that for expanded speech at 80 dB SPL. The error patterns obtained in these two conditions were different, suggesting that the differences between conditions in Experiment 1 were due largely to expanded amplitude envelopes rather than differences in audibility.

Consonant recognition for spectrally degraded speech as a function of consonant-vowel intensity ratio.

Normal-hearing subjects' recognition of spectrally degraded speech was evaluated under conditions in which the consonant-vowel (C-V) intensity ratio was modified. Subjects identified 22 consonants presented in an /a-Consonant-a/ format at suprathreshold and near threshold conditions. Prior to C-V ratio modification, the stimuli were processed to limit spectral information. In the suprathreshold experiment, stimuli were presented at the natural C-V ratio and at six modified C-V ratios. C-V ratio manipulations had dramatic effects on recognition scores for some groups of consonants. Scores for glides and sibilant fricatives increased with increasing C-V ratio, while scores for nasals and weak fricatives were best at low C-V ratios. Recognition scores for the stops and affricates were generally independent of C-V ratio. For nearly all consonants, changes in recognition scores as a function of C-V ratio were associated with changes in a general response bias toward those sounds. Performance was poor when the spectrally smeared VCVs were presented at low SN ratios in a second experiment. The influence of C-V ratio on percent correct scores was generally similar to the suprathreshold results, but consonant audibility, which was not a factor in the suprathreshold experiment, appeared to alter the effect of C-V ratio modification for some consonants.

Listeners' expectations about echoes can raise or lower echo threshold.

Echo threshold increases with exposure to redundant trains of stimuli. Three experiments were conducted to test the hypothesis that a change in the ongoing train would affect listeners' perception of the echo, but only if it signified an unusual change in room acoustics. The stimulus train was composed of 4-ms narrow-band noise bursts, with the leading sound from a loudspeaker placed 45 degrees left of midline and the lagging sound or simulated echo from 45 degrees right, delivered in an anechoic chamber. The lagging sound in the test noise, which followed the train after a 750-ms pause, came randomly from loudspeakers at 35 degrees or 55 degrees right, and the listener's task was to choose which position the echo came from on each trial. In experiment 1 the delay between onsets of the leading and lagging bursts was varied between train and test bursts, which simulated a sudden movement of the reflecting surface either toward the listener (if the delay of the test burst was shorter than the train) or away (if the delay was longer). In both cases listeners detected the echo's direction more easily, compared to trials when there was no change between train and test burst delays. In order to check whether any change between train and test bursts would increase echo discriminability, experiment 2 varied frequency and experiment 3 varied intensity. These variations were not expected to affect the echo's detectability because such changes signify that the original sound changed in these characteristics and the echo reflected these changes.(ABSTRACT TRUNCATED AT 250 WORDS)

Potential benefits of adaptive frequency-gain characteristics for speech reception in noise.

Some current single-microphone hearing aids employ techniques for adaptively varying the frequency-gain characteristics in an attempt to improve speech reception in noise. The potential benefit of this strategy depends on the spectral spread of masking and the degree to which it can be reduced by changing the frequency-gain characteristic. In this study these benefits were examined for subjects with normal hearing under static listening conditions. In the unprocessed condition, subjects were presented with nonsense syllables in an octave-band noise centered on 0.5, 1, or 2 kHz. The frequency-gain characteristic was then modified with the goal of reducing the intensity of the frequency region containing the octave-band noise. This processing resulted in increases as large as 60 percentage points in consonant-correct scores with the low- and mid-frequency octave noise bands, and a small increase with the high-frequency noise. Masking patterns produced by the octave noises were also measured and were related to the intelligibility results via an analysis based on Articulation Theory. The Articulation Index was also used to compare the effectiveness of three adaptive rules. A simple multiband volume control is expected to provide much of the benefit of more sophisticated systems without the need for separate estimation of input speech and noise spectra.

Frequency discrimination as a function of signal frequency and level in normal-hearing and hearing-impaired listeners.

Frequency difference limens (DLFs) for pure tones were obtained over a wide range of frequencies and levels from 7 normal-hearing subjects and 16 ears of 12 listeners with sensorineural hearing losses. The normal data were fitted with a general prediction equation. Variability of the data around the DLFs estimated by the equation was quantified and used to evaluate the DLFs from the hearing-impaired listeners. The majority of DLFs from impaired listeners were poorer than one standard deviation above the estimates of the normal equation at all frequencies and sensation levels (SLs). The portion of the equation concerned with sensation level was fitted to each listener's data at each frequency. The slopes of these functions indicated that, on average, the rate of improvement of the DLF with sensation level was similar in the two groups of subjects. These results suggest that it would be reasonable to compare DLFs from normal-hearing and hearing-impaired listeners at equivalent sensation levels. The intercepts of the DLF-intensity functions represent asymptotic values obtained at high SLs. These asymptotic DLFs were abnormal in the majority of hearing-impaired subjects, with more than half the data in excess of two standard deviations above normal. However, among those subjects, the correlation between the DLF deficit and the amount of hearing loss at the test frequency was not strong (r = +.27).

Dynamic processes in the precedence effect.

Three experiments were conducted to investigate the dependence of echo suppression on the auditory stimulation just prior to a test stimulus. Subjects sat in an anechoic chamber between two loudspeakers, one which presented the "lead" sound, and the other the delayed "lag" sound. In the first experiment, subjects reported whether or not they heard an echo coming from the vicinity of the lag loudspeaker during a test click pair. In seven of nine listeners, perception of the lagging sound was strongly diminished by the presence of a train of "conditioning" clicks presented just before the test click. Echo threshold increased (subjects were less sensitive to echoes) as the number of clicks in the train increased from 3 to 17. For a fixed number of clicks, the effect was essentially independent of click rate (from 1/s through 50/s) and duration of the train (from 0.5 through 8 s). A second experiment demonstrated a similar buildup of echo suppression with white noise bursts, regardless of whether the bursts in the conditioning train were repeated samples of frozen noise, or were independent samples of noise. Using an objective procedure for measuring echo threshold, the third experiment demonstrated that both lead and lag stimuli must be presented during the conditioning train in order to produce the buildup of suppression. When only the lead sound was presented during the conditioning train, the perceptibility of the lag sound during the test burst appeared to be enhanced.

Effect of consonant-vowel ratio modification on amplitude envelope cues for consonant recognition.

This investigation examined the degree to which modification of the consonant-vowel (C-V) intensity ratio affected consonant recognition under conditions in which listeners were forced to rely more heavily on waveform envelope cues than on spectral cues. The stimuli were 22 vowel-consonant-vowel utterances, which had been mixed at six different signal-to-noise ratios with white noise that had been modulated by the speech waveform envelope. The resulting waveforms preserved the gross speech envelope shape, but spectral cues were limited by the white-noise masking. In a second stimulus set, the consonant portion of each utterance was amplified by 10 dB. Sixteen subjects with normal hearing listened to the unmodified stimuli, and 16 listened to the amplified-consonant stimuli. Recognition performance was reduced in the amplified-consonant condition for some consonants, presumably because waveform envelope cues had been distorted. However, for other consonants, especially the voiced stops, consonant amplification improved recognition. Patterns of errors were altered for several consonant groups, including some that showed only small changes in recognition scores. The results indicate that when spectral cues are compromised, nonlinear amplification can alter waveform envelope cues for consonant recognition.

Effects of stimulus level on forward-masked psychophysical tuning curves in quiet and in noise.

Forward-masked psychophysical tuning curves were obtained from normal-hearing listeners at different probe levels in quiet and in a broadband background noise. In quiet, tuning-curve shape changed with probe level. For six listeners, tuning curves became broader with increasing probe level, primarily due to a decrease in the low-frequency slopes. For one listener, tuning curves became narrower with increasing probe level. The addition of a background noise, which was presented continuously at a level 10 dB below the noise level required to mask the probe tone, reduced the masker levels required to mask the probe tone. The reduction was greater near the tip of the tuning curve than on the tail, so that tuning curves in background noise were narrower than those obtained in quiet. Tuning curves with comparable masker levels near the tip of the tuning curve (Lmtip) were similar in shape, regardless of probe level or whether tuning curves were obtained in quiet or noise. Comparisons of tuning-curve characteristics derived by fitting tuning curves with least-squares procedures, indicated that low-frequency slopes decreased with Lmtip. As a consequence, Q10 dB values decreased with Lmtip. These results are consistent with the interpretation that tuning-curve shapes are determined by the intensities of the maskers required to mask the probe tone. The addition of a background noise restricted (partially masked) the excitation pattern of the probe so that lower masker intensities were required to "forward mask" the probe tone, and narrower tuning curves resulted from less intense markers. The results are well described by a two-process model of auditory excitation patterns.

The importance of consonant-vowel intensity ratio in the intelligibility of voiceless consonants.

The purpose of this study was to evaluate the extent to which variations in the consonant-vowel (C-V) intensity ratio could account for variations in speech intelligibility among the productions of 10 talkers. Fifty normal-hearing individuals listened in noise to syllables consisting of voiceless consonants followed by the vowel /a/ under three conditions in which: (a) C-V ratio varied naturally as produced by the talkers, and the stimuli were calibrated according to vowel intensity; (b) C-V ratios were increased and equated via digital signal processing; and (c) C-V ratios were unmodified, but the syllables were calibrated according to consonant level rather than vowel level. Results indicated that variations in C-V ratio explained a great deal of the variation in the intelligibility of some consonants (/s, S, tS/) but not others (the voiceless stops). This difference may well be due to differences in audibility between the two groups of consonants when they are presented at similar consonant-to-noise ratios. The majority of the data suggest that the importance of C-V ratio is related to the intensity of consonants but is independent of the ratio per se between consonant and vowel levels.

Effect of click rate and delay on breakdown of the precedence effect.

The precedence effect was tested as a function of echo-click delay and click rate after an abrupt switch in location between leading and lagging clicks. Click trains at three rates, 1/sec, 2/sec, and 4/sec, with delays ranging between 2 and 20 msec, were presented to subjects in an anechoic chamber. Duration of the click train after the switch in location was 12 sec, and echo click perceptibility was assessed throughout this period. The number of echo clicks heard was an increasing monotonic function of delay. The subjects reported a "fade-out" of echo clicks after a set number of clicks at each delay, regardless of rate. This result was interpreted as a buildup in inhibition of echoes produced by the ongoing click train. Suppression of echoes was stronger when the leading click originated from the right side than from the left side.

Temporal resolution in sensorineural hearing-impaired listeners.

Temporal masking curves were obtained from 12 normal-hearing and 16 hearing-impaired listeners using 200-ms, 1000-Hz pure-tone maskers and 20-ms, 1000-Hz fixed-level probe tones. For the delay times used here (greater than 40 ms), temporal masking curves obtained from both groups can be well described by an exponential function with a single level-independent time constant for each listener. Normal-hearing listeners demonstrated time constants that ranged between 37 and 67 ms, with a mean of 50 ms. Most hearing-impaired listeners, with significant hearing loss at the probe frequency, demonstrated longer time constants (range 58-114 ms) than those obtained from normal-hearing listeners. Time constants were found to grow exponentially with hearing loss according to the function tau = 52e0.011(HL), when the slope of the growth of masking is unity. The longest individual time constant was larger than normal by a factor of 2.3 for a hearing loss of 52 dB. The steep slopes of the growth of masking functions typically observed at long delay times in hearing-impaired listeners' data appear to be a direct result of longer time constants. When iterative fitting procedures included a slope parameter, the slopes of the growth of masking from normal-hearing listeners varied around unity, while those from hearing-impaired listeners tended to be less (flatter) than normal. Predictions from the results of these fixed-probe-level experiments are consistent with the results of previous fixed-masker-level experiments, and they indicate that deficiencies in the ability to detect sequential stimuli should be considerable in hearing-impaired listeners, partially because of extended time constants, but mostly because forward masking involves a recovery process that depends upon the sensory response evoked by the masking stimulus. Large sensitivity losses reduce the sensory response to high SPL maskers so that the recovery process is slower, much like the recovery process for low-level stimuli in normal-hearing listeners.

Frequency discrimination of short- versus long-duration tones by normal and hearing-impaired listeners.

This investigation explored the effects of stimulus level on the frequency discrimination of long- and short-duration pure tones by 5 subjects with normal hearing and 7 with sensorineural hearing impairment. Frequency difference limens (DLs) were obtained as a function of signal intensity for 5-ms and 300-ms tones at 500, 1000, and 2000 Hz. The performance of most of the hearing-impaired subjects was poorer than normal for 300-ms tones, but not for 5-ms tones. This result was relatively independent of the stimulus sensation levels at which the data were compared. However, the current results also show an unexpected dependence of the frequency DL on the sensation level of short-duration tones. In several normal-hearing subjects, frequency discrimination performance for these short tones is poorer at moderately high levels than at low levels.

Frequency discrimination as a function of tonal duration and excitation-pattern slopes in normal and hearing-impaired listeners.

Frequency difference limens were determined as a function of stimulus duration in five normal-hearing and seven hearing-impaired subjects. The frequency DL duration functions obtained from normal-hearing subjects were similar to those reported by Liang and Chistovich [Sov. Phys. Acoust. 6, 75-80 (1961)]. As duration increased, the DL's improved rapidly over a range of short durations, improved more gradually over a middle range of durations, and reached an asymptote around 200 ms. The functions obtained from the hearing-impaired subjects were similar to those from normal subjects over the middle and longer duration, but did not display the rapid changes at short durations. The paper examines the ability of a variation of Zwicker's excitation-pattern model of frequency discrimination to explain these duration effects. Most, although not all, of the effects can be adequately explained by the model.