Biomedical subjects
Publications and source records attributed to E Zwicker.
Sets of level and phase data of the (2f1-f2)-cancellation tone were measured with the same fit of the earphone. The results obtained were transferred into vector diagrams as x-y representations. Characteristic rules for the development of the cancellation tone can be extracted from a detailed analysis of such diagrams.
Using the Békésy tracking method, tone-on-tone masking patterns were measured with eight subjects at eight masker frequencies between 250 Hz and 6.03 kHz. Masking patterns were obtained for five masker levels between 20 and 60 dB SPL. Test-tone frequencies were chosen within two critical bands above and below the masker frequency. A level-dependent frequency asymmetry of masking was observed at all masker frequencies except at 250 Hz. At low levels there was a greater spread of masking towards the lower frequencies than toward the higher frequencies, but at high levels there was a greater spread of masking towards the higher frequencies than toward the lower frequencies. Only at masker levels of 40 dB SPL were the masking patterns approximately symmetrical.
Temporal resolution is, next to frequency resolution, the most important ability of the ear for perceiving speech. A simplified method for measuring temporal resolution clinically is described. It is based on masking period patterns, but measures only their maximum and minimum through the use of continuous and square-wave amplitude-modulated masking noises. Test tones and filtered masking noises are presented at 500, 1 500, and 4 000 Hz. Temporal resolution data of a normally hearing group and of seven pathological groups are given: Patients with conductive hearing loss show normal resolution; those with noise-induced, age-induced, and sudden hearing loss show normal resolution at low and middle frequencies, but reduced resolution at 4 000 Hz; patients with retrocochlear hearing loss and Menière's disorder show a tendency for reduced resolution at all frequencies, but reduction at 4 000 Hz, while ototoxic hearing loss shows the greatest reduction at 1 500 Hz. The method used also gives an indication for reduced frequency selectivity.
A suppression-period pattern' for acoustical responses has been measured in a similar way to the masking-period pattern. A triggered low-frequency masker decreases the acoustical response to a short test signal differently at different presentations within the period of the masker. The two patterns show such close relationships that a masking-period pattern can be developed out of a set of suppression-period patterns: masked threshold is reached with the same parameter conditions under which the acoustic responses diminish. The rules elaborated for masking-period patterns seem to hold for suppression-period patterns as well.
Acoustic responses to short sound stimuli have been measured in the external ear canal of guinea pigs. Unequivocal responses had frequencies between 1.8 and 3.2 kHz, latencies (to their peak pressures) of 2.8 to 4.5 ms and sound pressures up to -10 dB SPL. A set of criteria was developed to clearly distinguish acoustical responses from after-resonances of the stimuli. Added phase-locked, very-low-frequency tones were able under some conditions to totally suppress the acoustical responses. The suppression-period pattern thus produced strongly resemble those obtained for man and the masking-period patterns obtained with human subjects.
Explore the source record for details and available documents.
A rough approximation of the saturation-feedback model is used to calculate the dependence of the (2f1 - f2) difference tone on the level L1 of the lower primary. The calculated data are in astonishingly good agreement with measured data, indicating that the cubic difference tone is a vector sum composed from many partials. These partials are derived through a nonlinear saturation characteristic with a knee point corresponding to 40 dB SPL from the excitation-time functions produced by the two primaries along the basilar membrane.
This study compares frequency selectivity--as measured by four different methods--in observers with normal hearing and in observers with conductive (nonotosclerotic), otosclerotic, noise-induced, or degenerative hearing losses. Each category of loss was represented by a group of 7 to 10 observers, who were tested at center frequencies of 500 Hz and 4000 Hz. For each group, the following four measurements were made: psychoacoustical tuning curves, narrow-band masking, two-tone masking, and loudness summation. Results showed that (a) frequency selectivity was reduced at frequencies where a cochlear hearing loss was present, (b) frequency selectivity was reduced regardless of the test level at which normally-hearing observers and observers with cochlear impairment were compared, (c) all four measures of frequency selectivity were significantly correlated and (d) reduced frequency selectivity was positively correlated with the amount of cochlear hearing loss.
Recent misinterpretations [Nelson, 1979] of an earlier paper [Zwicker, 1954] are corrected and measurements were repeated in more detail to show the reversal of masking with masker level more clearly: masking extends predominantly towards high frequencies for high levels; it extends, however, more towards low frequencies for low levels.
Data from literature related to nonlinearities of the peripheral part of the hearing system are collected and extended by results from measurements of acoustical responses, masking, cubic difference tones and Zwicker tones. The data indicate 40 dB as a significant value for the dynamic range in neurophysiology as well as for the sensation level in psychoacoustics dividing the total level range into two areas of different characteristics. A preliminary model assuming that the outer hair cells act as an amplifier which contains saturation (corresponding to 40 dB) and feed back to sensitize the inner hair cells is used to describe the measured effects at least qualitatively.
Level and phase of the cubic and quadratic difference tone were measured as a function of the primary tones' level, frequency separation, frequency region as well as of the frequency of an additional simultaneous masker. The results lead to the conclusion that quadratic and cubic difference tones are produced in different nonlinearities although both result in the same kind of adequate stimulus.
The main contention of this paper is that Zwicker's model of loudness summation is applicable to observers with noise-induced hearing loss when certain parameters of the model are modified. Two types of measurement were obtained in observers with normal hearing and noise-induced hearing loss: loudness summation as a function of level and narrow-band masking. These measurements provided a basis for modifying the parameters of the model. Results suggest that the model of loudness summation is applicable to observers with noise-induced hearing loss when the presence of recruitment and reduced frequency selectivity is taken into account.
An approach to automatic speech recognition is described, which, in a straightforward way, follows the concept of (1) preprocessing in terms of auditory parameters and (2) subsequent classification and recognition. The preprocessing system has been realized in analog hardware, while recognition is carried out on a digital computer. In the preprocessing system, the essential psychoacoustic principles of the perception of loudness, pitch, roughness, and subjective duration are implemented with some approximation. The system essentially consists of 24 bandpass filters, nonlinear transformation of each filter output into specific loudness and specific roughness, and final transformation of these parameters into total loudness, total roughness, and three spectral momenta. As a means to further reduce the information flow, continuous selection of dominant parameters is also considered on the basis of psychoacoustic data. The subsequent recognition process is mainly characterized by (1) discrimination between speech and silent periods, (2) detection of syllable peaks and classification of syllable nuclei, and (3) assumption of syllable boundaries and classification of consonant clusters. Though the entire system as yet is far from being complete and perfect, the present results indicate that the concept provides a systematic and promising way towards automatic recognition of continuous speech.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
Explore the source record for details and available documents.
An apparatus for clinical use is described by which simplified psychoacoustical tuning curves can be measured near 500 Hz and near 4 kHz. Data produced by four groups (normal hearing, conductive hearing loss, noise induced hearing loss and degenerative hearing loss) are separately averaged and compared. The first mentioned two groups as well as the third in the low frequency range show good frequency resolution, the others, however, show largely decreased frequency resolution.
Explore the source record for details and available documents.