Phase independence of pitch produced by narrow-band sounds.

Three listeners matched the pitch of a simple tone to that of narrow-band complex signals having different phases. The pitch matches were independent of the phases; the frequency of the simple tone approximately equaled the center of gravity of the power spectrum of each complex signal. This result is inconsistent with a model that calculates the pitch of a waveform as the average of instantaneous frequency weighted by the envelope of the waveform.

Intensity-weighted average of instantaneous frequency as a model for frequency discrimination.

The intensity-weighted average of instantaneous frequency (IWAIF) is developed as a model to predict listener performance in tasks primarily requiring frequency discrimination. IWAIF is closely related to the envelope weighted average of instantaneous frequency (EWAIF) model proposed by Feth for similar tasks. The primary difference is that the IWAIF model uses intensity (envelope squared) as the weighting function instead of the envelope. The advantages of IWAIF over EWAIF are that (a) it has a convenient frequency domain interpretation; and (b) it is much simpler to compute than the EWAIF. The IWAIF is the "center of gravity" of the energy spectral density function of the signal.

Temporal resolution in normal-hearing and hearing-impaired listeners using frequency-modulated stimuli.

This study compares the temporal resolution of frequency-modulated sinusoids by normal-hearing and hearing-impaired subjects in a discrimination task. One signal increased linearly by 200 Hz in 50 msec. The other was identical except that its trajectory followed a series of discrete steps. Center frequencies were 500, 1000, 2000, and 4000 Hz. As the number of steps was increased, the duration of the individual steps decreased, and the subjects' discrimination performance monotonically decreased to chance. It was hypothesized that the listeners could not temporally resolve the trajectory of the step signals at short step durations. At equal sensation levels, and at equal sound pressure levels, temporal resolution was significantly reduced for the impaired subjects. The difference between groups was smaller in the equal sound pressure level condition. Performance was much poorer at 4000 Hz than at the other test frequencies in all conditions because of poorer frequency discrimination at that frequency.

Effects of intense pure tones on auditory temporal acuity.

Gap detection thresholds (GDTs) were obtained from human listeners before and after exposure to a brief 0.4- or 1.7-kHz tone. The temporary threshold shift (TTS) produced 2 min after an exposure was approximately 10 dB. GDT stimuli were octave-band noises centered at one of three frequencies: the exposure frequency, one-half octave above the exposure frequency or one octave above the exposure frequency. GDTs were obtained at 35, 55, and 75 dB SPL at each center frequency. GDT and TTS recovery were monitored at logarithmically-spaced time intervals after the exposures. Following the 1.7-kHz exposure, shifts in post-exposure GDT were only obtained with the low-level stimulus conditions--the magnitude of GDT shift was correlated with the size of the TTS, and the shifts in GDT and absolute threshold required similar amounts of time to recover. Significant post-exposure shifts in GDT were also observed following the 0.4-kHz exposure. However, shifts were found at frequencies where there was no measurable TTS, and they required longer periods of time to recover than did absolute threshold.

The relation between the human frequency-following response and the low pitch of complex tones.

The relation between the auditory brain stem potential called the frequency-following response (FFR) and the low pitch of complex tones was investigated. Eleven complex stimuli were synthesized such that frequency content varied but waveform envelope periodicity was constant. This was accomplished by repeatedly shifting the components of a harmonic complex tone upward in frequency by delta f of 20 Hz, producing a series of six-component inharmonic complex tones with constant intercomponent spacing of 200 Hz. Pitch-shift functions were derived from pitch matches for these stimuli to a comparison pure tone for each of four normal hearing adults with extensive musical training. The FFRs were recorded for the complex stimuli that were judged most divergent in pitch by each subject and for pure-tone signals that were judged equal in pitch to these complex stimuli. Spectral analyses suggested that the spectral content of the FFRs elicited by the complex stimuli did not vary consistently with component frequency or the first effect of pitch shift. Furthermore, complex and pure-tone signals judged equal in pitch did not elicit FFRs of similar spectral content.

Effects of normal and pathologic eardrum impedance on sound pressure in the aided ear canal: a computer simulation.

Reported herein are results of computer simulations of aided sound spectra in ears with normal and pathologic eardrum impedance. The computer technique used in this study has been reported elsewhere [D. P. Egolf, D. R. Tree, and L. L. Feth, J. Acoust. Soc. Am. 63, 264-271 (1978)]. Consequently, to develop reader confidence in the computer scheme, its application to real ears was first tested. This was accomplished by (1) comparing computed spectral data with in-the-ear measurements and (2) comparing real ear minus 2-cc coupler data-both computer generated--with an idealized difference curve published elsewhere [R. M. Sachs and M. D. Burkhard, unpublished rep. no. 20022-1, Industrial Research Products, Inc., Elk Grove Village, IL (1972)]. Results indicate that the wide variation in eardrum impedance among normals evidenced in other studies produces a corresponding wide variation in aided spectrum shape. Likewise, simulations utilizing two sets of pathologic eardrum impedance data obtained from the literature show that aided sound spectra in such ears are likely to be significantly different from those occurring in normal ears. These findings suggest, as others have concluded, that there may be a substantial variation in spectrum shape among individuals wearing identically the same hearing aid--even if those individuals have normal hearing. In conclusion, questions are raised about the use of real-ear simulators and the need for a comprehensive computer-based model of an entire hearing aid.

Temporal integration of forward masking in listeners having sensorineural hearing loss.

Findings are described from three experiments that were designed to contrast estimates of auditory time and frequency analysis in highly practiced, young adult subjects having sensorineural hearing loss of cochlear origin with subjects having normal hearing. In one experiment, pure-tone detection threshold was measured for two durations of tones. For the hearing-impaired subjects, smaller differences were found between thresholds for the two durations of tones than were found for the normally hearing subjects at test frequencies where the sensitivity loss was greatest. In a subsequent forward masking experiment, masked thresholds were obtained at 3 kHz as a function of masker level at several masker durations. The results indicated that, for the range of values tested, masked probe thresholds changed less as a function of masker duration for the hearing-impaired subjects than for normally hearing subjects. However, forward masking grew as a function of masker level more steeply for the hearing-impaired listeners than for the normally hearing listeners. We believe that this result indicates abnormal temporal processing of sequential sounds. In a third experiment, psychophysical tuning curves were measured in a forward masking procedure for two maskers that differed in duration by approximately an order of magnitude. Again, the hearing-impaired subjects demonstrated less change in forward masking as a function of masker duration than normally hearing subjects. Further, the sharpness of the tuning curves from the hearing-impaired listeners was markedly reduced as compared to normal. It is believed that the results reflect a disruption of the normal temporal and spectral representation of sounds in the hearing-impaired subjects.

Characteristics of the glottal turbulent noise source.

This investigation utilized a reflectionless tube technique to obtain direct estimates of turbulent noise produced at the glottis during whispered vowels. In the past, the glottal turbulent noise source has been described theoretically as a series pressure source having a spectrum that is relatively flat for 2 or 3 oct around a center frequency [K.N. Stevens, J. Acoust. Soc Am. 50, 1180-1192 (1971)]. Center frequency is determined primarily by the area of the constriction at which turbulence is produced with the volume velocity of air flowing through the constriction. The present results were shown to substantiate this theoretically based description of the glottal turbulent noise source. In addition, there was no significant difference between the glottal turbulent noise spectra of male and female speakers. The application of these findings to the synthesis of whispered vowels is discussed.

Pitch of narrow-band signals.

This study investigated the pitch elicited by complex narrow bandwidth signals. These signals ranged from two-component tones to multiple-component approximations of narrow-band noise. All were contained within 10-, 20-, or 50-Hz bandwidths. Listeners were asked to adjust the frequency of a pure tone to match the pitch they heard in a given complex signal. A simple model suggests that the pitch of these complex signals should match that of a pure tone set to the center frequency. For a majority of the signals, the performance of three of our four listeners was not different from the model predictions. However, the two-component signals were apparently resolved and two simultaneous pitches were heard by some listeners. Our fourth listener heard two sequential pitches in many of the complex signals. We are unable to account for that performance although the pitch matches were very repeatable. These results have implications for an understanding of pitch coding and auditory spectral resolution.

Hearing aid microphone location effects on speech discrimination in noise.

The interaction of the head with a sound impinging upon it has a direct effect on the sound as it is seen at the port of the hearing aid microphone. While other investigators have evaluated this effect in terms of changes in the frequency response of the hearing aid, this investigation sought to evaluate the significance of the effect in terms of the intelligibility of speech presented in a noisy background. The three typical locations of a hearing aid microphone were simulated with a high-fidelity probe-tube microphone placed around the right ear of KEMAR. The locations were: over the ear, behind the ear, and in the ear. An earmold was kept in the ear at all times. Speech and noise were presented to the microphone and recorded on tape for presentation to normally hearing subjects. The results indicated that so long as the hearing aid microphone is located on the head, around the ear, no one location is better than any other for speech intelligibility.

Patterns of residual masking.

"Residual masking' was measured in a tonal forward masking paradigm. In one experiment, psychophysical tuning curves and masking patterns were obtained at several frequencies and levels for a fixed masker-probe time delay. In a second experiment, tuning curves and masking patterns were measured as masker-probe time delay was varied. Our results indicate that tuning curves and masking patterns are sharpest at low levels, high frequencies and brief masker-probe time delays. In addition, we observed that masked probe threshold returned to the level of unmasked probe threshold at approximately the same post-masker time regardless of masker level or the probe-to-masker frequency relationship. These findings suggest that frequency, level and time delay all affect the degree of frequency selectivity observed with these measures.

Relationship between psychophysical tuning curves and "suppression".

This study examined two-tone unmasking and auditory frequency selectivity about 3 kHz for the purpose of demonstrating a qualitative relationship between the two. An adaptive 2IFC forward-masking procedure was used to collect psychophysical tuning curves (PTC's) and two-tone masking data under a quiet and noise condition for the same normal-hearing listeners. In the noise condition, a narrowband noise masker, centered one decade down from the probe, was gated on with the tonal masker(s). Kiang and Moxon [J. Acoust. Soc. Am. 55, 620-630 (1974)] have found that low-frequency narrowband noise serves to decrease the sharpness of electrophysiological tuning curves by affecting only the tip segments. The data for four highly practiced listeners indicate that the gated-noise masker was effective in broadening the PTC's and in lessening the magnitude of two-tone unmasking. The mutually reflected changes in tuning curves and in two-tone unmasking indicate a close relationship between frequency selectivity and unmasking: the greater the magnitude of unmasking above the center frequency of the PTC, the sharper the tuning of the PTC.

Vocal roughness and jitter characteristics of vowels produced by esophageal speakers.

Audiotape recordings of sustained vowels produced by nine esophageal speakers were subjected to acoustic and perceptual analysis. Results indicated that (1) the magnitude of vocal jitter present in the vowels was substantially larger than that observed in normal speakers and speakers with laryngeal/vocal disturbance, (2) listeners could reliably rate the severity of vocal roughness in the vowels, (3) voices of esophageal speakers were characterized by varying degrees of vocal roughness, and (4) mean fundamental frequency, mean jitter, or jitter ratio measures did not serve as useful predictors of the perceived severity of vocal roughness. These findings are interpreted to suggest that the mechanism esophageal speakers employ to regulate fundamental frequency is substantially different from that employed by normal speakers and that the identity of physical variables underlying the perception of roughness severity in naturally produced human speech is not well understood.

Mathematical predictions of electroacoustic frequency response of in situ hearing aids.

The amplitude spectrum of an acoustic signal presented to the microphone of a hearing aid is altered drastically before it finally reaches the user's eardrum. A major part of this alteration is due to the interaction of various mechanical and acoustic resonances which are characteristic of the hearing-aid receiver and the sound transmission system linking the receiver with the eardrum. Because of the complexity of this phenomenon, there is yet no means for predicting, a priori, the true shape of the sound spectrum that will occur at the user's eardrum. This paper reports on the development and testing of just such a scheme. The accuracy of this scheme--a computer-aided mathematical technique--is measured in the laboratory on real and artificial ears. The results of those measurements show good agreement between experimental and computer-generated data below 5000 Hz.

Two-tone auditory spectral resolution.

Two component complex tones were synthesized so that each member of a complementary pair exhibited identical amplitude modulations; frequency modulations, however, were in opposite directions. Previous work has shown that the discriminability of the members of a complementary pair is related to the envelope-weighted differences in their instantaneous frequency functions. Listeners report that they base their discriminations upon differences in pitch between complementary two component signals. We assumed that the high discriminability found for moderate separations between the two component frequencies would not obtain if the components were resolved into separate critical bands. Thus we could define the just discriminable (75%) frequency separation as a direct estimate of spectral resolving power without recourse to assumptions about masking effectiveness, loudness summation, or other subjective changes thought to depend upon auditory spectral resolving power. Our findings indicate that resolution bandwidths resulting from the discriminability of complex tone pairs approximate the width of the traveling-wave envelope observed by von Békèsy. In light of our results, the implications for other critical bandwidth estimates are discussed.