Masking In Listeners Research Articles

Binaural sensitivity and release from speech-on-speech masking in listeners with and without hearing loss.

Listeners with sensorineural hearing loss routinely experience less spatial release from masking (SRM) in speech mixtures than listeners with normal hearing. Hearing-impaired listeners have also been shown to have degraded temporal fine structure (TFS) sensitivity, a consequence of which is degraded access to interaural time differences (ITDs) contained in the TFS. Since these "binaural TFS" cues are critical for spatial hearing, it has been hypothesized that degraded binaural TFS sensitivity accounts for the limited SRM experienced by hearing-impaired listeners. In this study, speech stimuli were noise-vocoded using carriers that were systematically decorrelated across the left and right ears, thus simulating degraded binaural TFS sensitivity. Both (1) ITD sensitivity in quiet and (2) SRM in speech mixtures spatialized using ITDs (or binaural release from masking; BRM) were measured as a function of TFS interaural decorrelation in young normal-hearing and hearing-impaired listeners. This allowed for the examination of the relationship between ITD sensitivity and BRM over a wide range of ITD thresholds. This paper found that, for a given ITD sensitivity, hearing-impaired listeners experienced less BRM than normal-hearing listeners, suggesting that binaural TFS sensitivity can account for only a modest portion of the BRM deficit in hearing-impaired listeners. However, substantial individual variability was observed.

Determining the energetic and informational components of speech-on-speech masking in listeners with sensorineural hearing loss.

The ability to identify the words spoken by one talker masked by two or four competing talkers was tested in young-adult listeners with sensorineural hearing loss (SNHL). In a reference/baseline condition, masking speech was colocated with target speech, target and masker talkers were female, and the masker was intelligible. Three comparison conditions included replacing female masker talkers with males, time-reversal of masker speech, and spatial separation of sources. All three variables produced significant release from masking. To emulate energetic masking (EM), stimuli were subjected to ideal time-frequency segregation retaining only the time-frequency units where target energy exceeded masker energy. Subjects were then tested with these resynthesized "glimpsed stimuli." For either two or four maskers, thresholds only varied about 3 dB across conditions suggesting that EM was roughly equal. Compared to normal-hearing listeners from an earlier study [Kidd, Mason, Swaminathan, Roverud, Clayton, and Best, J. Acoust. Soc. Am. 140, 132-144 (2016)], SNHL listeners demonstrated both greater energetic and informational masking as well as higher glimpsed thresholds. Individual differences were correlated across masking release conditions suggesting that listeners could be categorized according to their general ability to solve the task. Overall, both peripheral and central factors appear to contribute to the higher thresholds for SNHL listeners.

Spatial release from masking in listeners with asymmetric hearing thresholds

Spatial release from masking was evaluated as a function of target-masker spatial separation in listeners with normal hearing and with chronic asymmetric hearing loss. The target was one of 20 vocoded words spoken by a female talker. The masker was a stream of vocoded nonsense sentences spoken by two other female talkers. Subjects listened in an anechoic chamber with loudspeakers at 1.9 m distance. The target was always presented from directly in front while the masker was presented from the front or from left or right at varying angular separations. In some conditions a copy of the masker with a delay of 4 ms was introduced at the front loudspeaker, simulating a reflection. Subjects detected the target in an adaptive 4AFC task. Results showed strong indications of informational masking in the co-located condition and spatial release that increased with masker angular separation, reaching almost 27 dB at wide separations in normal hearing listeners, and 22 dB even with the reflection. Listeners with asymm...

Spatial release from masking as a function of the spectral overlap of competing talkers

This study tested the hypothesis that the reduced spatial release from speech-on-speech masking typically observed in listeners with sensorineural hearing loss results from increased energetic masking. Target sentences were presented simultaneously with a speech masker, and the spectral overlap between the pair (and hence the energetic masking) was systematically varied. The results are consistent with increased energetic masking in listeners with hearing loss that limits performance when listening in speech mixtures. However, listeners with hearing loss did not exhibit reduced spatial release from masking when stimuli were filtered into narrow bands.

Robust Text-independent Speaker Identification in a Time-varying Noisy Environment

Practical speaker recognition systems are often subject to noise or distortions within the input speech which degrades performance. In this paper, we proposed a new mel-frequency cepstral coefficients (MFCC) based speaker identification system with Vector Quantization (VQ) modeling technique. It integrates a hearing masking effect based masker and a group of dozen triflers into traditional MFCC feature extraction for robust speaker identification. The masker can decrease the influence of noise signal to the speech signal, and improve the recognition rate. The mixture of triflers can enhance high-frequency calculation accuracy. A purposeful voice samples database are collected under an unconstrained indoor environment for a month. The texts to be spoken are also unconstrained. The proposed method is evaluated with the voice samples database, and its recognition rates remain over 93% under different experiment condition. The experiments results show that the proposed speaker identification system has good accuracy and robustness to the unconstrained noisy condition and text-independent.

Recovery from on- and off-frequency forward masking in listeners with normal and impaired hearing

The aim of this study was to investigate the possible mechanisms underlying an effect reported earlier [Wojtczak, M., and Oxenham, A. J. (2009). J. Acoust. Soc. Am. 125, 270-281] in normal-hearing listeners, whereby recovery from forward masking can be slower for off-frequency tonal maskers than for on-frequency tonal maskers that produce the same amount of masking at a 0-ms masker-signal delay. To rule out potential effects of confusion between the tonal signal and tonal masker, one condition used a noise-band forward masker. To test whether the effect involved temporal build-up, another condition used a short-duration (30-ms) forward masker. To test whether the effect is dependent on normal cochlear function, conditions were tested in five listeners with sensorineural hearing loss. For the 150-ms noise maskers, the data from normal-hearing listeners replicated the findings from the previous study that used tonal maskers. In contrast, no significant difference in recovery from on- and off-frequency masking was observed for the 30-ms tonal maskers in normal-hearing listeners, or for the 150-ms tonal maskers in hearing-impaired listeners. Overall, the results are consistent with a mechanism based on efferent feedback that affects the recovery from forward masking in the normal auditory system.

Speech production modifications produced in the presence of low-pass and high-pass filtered noise

In the presence of noise, do speakers actively shift their spectral energy distribution to regions least affected by the noise? The current study measured speech level, fundamental frequency, first formant frequency, and spectral center of gravity for read speech produced in the presence of low- and high-pass filtered noise. In both filtering conditions, these acoustic parameters increased relative to speech produced in quiet, a response which creates a release from masking for listeners in the low-pass condition but which actually increases masking in the high-pass noise condition. These results suggest that, at least for read speech, speakers do not adopt production strategies in noise which optimize listeners' information reception but that instead the observed shifts could be a passive response which creates a fortuitous masking release in the low-pass noise. Independent variation in parameters such as F0, F1 and spectral center of gravity may be severely constrained by the increase in vocal effort which accompanies Lombard speech.

Informational masking in listeners with sensorineural hearing loss.

Measures of energetic and informational masking were obtained from 46 listeners with sensorineural hearing loss. The task was to detect the presence of a sequence of eight contiguous 60-ms bursts of a pure tone embedded in masker bursts that were played synchronously with the signal. The masker was either a sequence of Gaussian noise bursts (energetic masker) or a sequence of random-frequency 2-tone bursts (informational masker). The 2-tone maskers were of two types: one type that normally tends to produce large amounts of informational masking and a second type that normally tends to produce very little informational masking. The two informational maskers are called "multiple-bursts same" (MBS), because the same frequency components are present in each burst of a sequence, and "multiple-bursts different" (MBD), because different frequency components are presented in each burst of a sequence. The difference in masking observed for these two maskers is thought to occur because the signal perceptually segregates from the masker in the MBD condition but fuses with the masker in MBS. In the present study, the effectiveness of the MBD masker, measured as the signal-to-masker ratio at masked threshold, increased with increasing hearing loss. In contrast, the signal-to-masker ratio at masked threshold for the MBS masker changed much less as a function of hearing loss. These results suggest that sensorineural hearing loss interferes with the ability of the listener to perceptually segregate individual components of complex sounds. The results from the energetic masking condition, which included critical ratio estimates for all listeners and auditory filter characteristics for a subset of the listeners, indicated that increasing hearing loss also reduced frequency selectivity at the signal frequency. Overall, these results suggest that the increased susceptibility to masking observed in listeners with sensorineural hearing loss is a consequence of both peripheral and central processes.

A spectral layer model in speech and music perception

According to a model of the peripheral hearing system establishing a masking and overmasking paradigm, speech and music signals have been split into several spectral layers by means of spectral subtraction. Spectral components below the so-called psychoacoustical irrelevance threshold are eliminated in a first processing step. The resulting signal, which exhibits no noticeable difference to the unprocessed one, is fed into an overmasking procedure applying flattened hearing masking functions. This action implements a kind of a spectral peak detector device. Consequently, the discrimination between spectral peaks (auditory figure) and background components, especially in reverberant environments, is improved. As a result, different audible signal parts according to different spectral layers are obtained, separating spectral peaks and weaker spectral components. In a certain range, both parts of speech and music signals are equally intelligible. In music, leading voices can be extracted and separated from the orchestra sound. Recent research indicates that this type of figure-background discrimination based on the dynamic characteristics of the acoustic signal only can exhibit several perceptually relevant features in listeners with normal hearing as well as in subjects with sensineural hearing impairment.

Growth of masking in listeners with cochlear impairments and impairments simulated by masking

The higher‐than‐normal masked thresholds observed in many impaired listeners may be explained by nonlinear additivity of masking by the external masker and an hypothetical internal masker [Jesteadt et al., J. Acoust. Soc. Am. Suppl. 1 81, S77 (1987); Humes et al., J. Acoust. Soc. Am. Suppl. 1 81, S77 (1987)]. To test this hypothesis, Békésy tracking was used to measure growth‐of‐masking functions for 250‐ms tones with frequencies at and above those of continuous low‐pass or octave‐band noise maskers. Results for four impaired listeners show that the minimum‐effective‐masker levels (MEMLs, above which masking occurs) are higher than normal for small masker‐signal separations, Δf's, but approach normal values for larger Δf's. Above the MEMLs, masking grows 1 dB/dB or less. Results for simulated impairments are similar to those for real impairments, but MEMLs tend to be higher and masking often is less in simulated than in real impairments. These results indicate that nonlinear additivity of internal and external masking accounts for some, but not all, of the excessive spread of masking observed in many impaired listeners. [Work supported by NIH.]

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.

Susceptibility to intraspeech spread of masking in listeners with sensorineural hearing loss.

Previous research with speechlike signals has suggested that upward spread of masking from the first formant (F 1) may interfere with the identification of place of articulation information signaled by changes in the upper formants. This suggestion was tested by presenting two-formant stop consonant--vowel syllables varying along a/ba--/da/--/ga/ continuum to hearing-impaired listeners grouped according to etiological basis of the disorder. The syllables were presented monaurally at 80 dB and 100 dB SPL when formant amplitudes were equal and when F 1 amplitude was reduced by 6, 12, and 18 dB. Noise-on-tone masking patterns were also generated using narrow bands of noise at 80 and 100 dB SPL to assess the extent of upward spread of masking. Upward spread of masking could be demonstrated in both speech and nonspeech tasks, irrespective of the subject's age, audiometric configuration, or etiology of hearing impairment. Attenuation of F 1 had different effects on phonetic identification in different subject groups: While listeners with noise-induced hearing loss showed substantial improvement in identifying place of articulation, upward spread of masking did not consistently account for poor place identification in other types of sensorineural hearing impairment.

Masking and partial masking in listeners with a high-frequency hearing loss.

3 listeners with sensorineural hearing loss ranging from moderate to moderate-severe starting at frequencies higher than 1 kHz participated in two masking experiments and a partial masking experiment. In the first masking experiment, fM = 1 KHz and LM = 50 dB SPL, higher than normal masked thresholds were obtained for listeners whose hearing was impaired in the frequency region of clear hearing loss as well as in the region of near-normal absolute thresholds. The second masking experiment showed that for hearing-impaired listeners the elevation of the masked thresholds, in decibels, in this frequency region of "near-normal' absolute thresholds was equal to the elevation of the absolute thresholds, in decibels. The third experiment, a partial masking experiment with fM = 975-1025 Hz and LM = 75 dB SPL, showed similar partial-masking functions for hearing-impaired and normal listeners, but the functions for the hearing-impaired listeners were at much higher levels of the partially masked probe tone. Thus the higher masked thresholds of the hearing-impaired can result in a dramatic reduction of the dynamic range of hearing under masking in the frequency region of the hearing loss and also in the region with only a small hearing loss (less than 30 dB). It is suggested that this may explain the speech perception difficulties which these listeners experience, especially in the presence of ambient noise.

Backward and forward masking in listeners with severe sensorineural hearing loss.

Using synthetic stimuli, backward and forward masking by a first-formant stimulus were measured for two tasks: (1) discrimination of a second-formant transition and (2) pure-tone detection. 16 listeners with moderate to severe sensorineural hearing impairment were studied. Many of the listeners showed backward and forward discrimination masking, sometimes extending as far as 200 ms before or after the first-formant interval. There were large individual differences in degree and temporal extent of masking. The masking contribution of sensitivity shifts, as indicated by pure-tone detection levels, appeared to be limited to a range of only 50 ms before or after the first-formant masker.

Pure-tone octave masking in listeners with sensorineural hearing loss.

Pure-tone octave masking was investigated in 14 listeners with sensorineural hearing loss to examine the hypothesis that the sensorineural ear introduces abnormal harmonic distortion. Thresholds for a test signal at f 2 , masked by a masking signal at f 1 , (where f 2 = 2f 1 ) were obtained as a function of the level of the f 1 masker for four different f 1 frequencies (250, 500, 1000, and 2000 Hz) and two different phase relations between the f 1 and f 2 signals (90° and 270°). Analysis of the data in terms of the absolute level of the f 2 test signal at masked threshold vs the absolute level of the f 1 masking signal leads to the conclusion that these pathological ears do not perform differently from normal ears, except along the dimension of hearing loss. That is, their hearing losses do not add significant distortion to the acoustic signal. Analysis of the data in terms of the sensation level of the f 2 test signal at masked threshold leads to the specious conclusion that the sensorineural ear introduces abnormal distortion.

Pure-Tone Octave Masking in Listeners with Sensorineural Hearing Loss

Phase-locked pure-tone masking was investigated in 16 listeners with sensorineural hearing loss to examine the hypothesis that the sensorineurai ear introduces abnormal distortion. Thresholds for a pure tone at f2, masked by a pure tone at f1 (where f2=2f1), were obtained as a function of the level of the f1 masker for three different f1 frequencies (250, 500, and 1000 Hz) and two different phase relations between f1 and f2 (90° and 270°). Estimates of the level of the aural harmonic were obtained from those masking functions. Analysis of the data in terms of the absolute level of f2 at masked threshold versus the absolute level of f1 results in a conclusion that these pathological ears do not perform differently from normal ears, except along the dimension of hearing loss. That is, their hearing loss does not add significant distortion to the acoustic signal. Even though the present analysis suggests that evaluation of the data for pure-tone octave masking in terms of SL leads to the specious conclusion that the sensorineural ear introduces abnormal distortion, an attempt was made to relate hearing loss to the aural harmonic in less simplistic terms [Research supported by grants from the NINDS of USPHS.]