Consonant Recognition Scores Research Articles

Preliminary evaluation of computer-assisted home training for French cochlear implant recipients.

For French cochlear implant (CI) recipients, in-person clinical auditory rehabilitation is typically provided during the first few years post-implantation. However, this is often inconvenient, it requires substantial time resources and can be problematic when appointments are unavailable. In response, we developed a computer-based home training software ("French AngelSound™") for French CI recipients. We recently conducted a pilot study to evaluate the newly developed French AngelSound™ in 15 CI recipients (5 unilateral, 5 bilateral, 5 bimodal). Outcome measures included phoneme recognition in quiet and sentence recognition in noise. Unilateral CI users were tested with the CI alone. Bilateral CI users were tested with each CI ear alone to determine the poorer ear to be trained, as well as with both ears (binaural performance). Bimodal CI users were tested with the CI ear alone, and with the contralateral hearing aid (binaural performance). Participants trained at home over a one-month period (10 hours total). Phonemic contrast training was used; the level of difficulty ranged from phoneme discrimination in quiet to phoneme identification in multi-talker babble. Unilateral and bimodal CI users trained with the CI alone; bilateral CI users trained with the poorer ear alone. Outcomes were measured before training (pre-training), immediately after training was completed (post-training), and one month after training was stopped (follow-up). For all participants, post-training CI-only vowel and consonant recognition scores significantly improved after phoneme training with the CI ear alone. For bilateral and bimodal CI users, binaural vowel and consonant recognition scores also significantly improved after training with a single CI ear. Follow-up measures showed that training benefits were largely retained. These preliminary data suggest that the phonemic contrast training in French AngelSound™ may significantly benefit French CI recipients and may complement clinical auditory rehabilitation, especially when in-person visits are not possible.

Speech perception abilities of adult cochlear implant listeners with single-sided deafness vs. bilateral hearing loss

Objectives The purpose of this study was to compare the speech perception abilities in adult cochlear implant recipients implanted for bilateral sensorineural hearing loss (BSNHL) with those implanted for single-sided deafness (SSD). Design A total of 12 adults with BSNHL and 12 adults with SSD participated. Each participant completed a battery of speech perception measures including monosyllabic words, sentences, and consonant recognition. Results Cochlear implant users with BSNHL performed higher on word and sentence recognition. Consonant recognition scores showed higher performance for CI listeners with BSNHL for voicing and manner, but not for place or articulation. Conclusions Results of this study suggest that adults with SSD may have lower speech perception abilities with their cochlear implant when compared to adults implanted for BSNHL.

Speech perception in noise in children with dyslexia: Does speech sound disorder matter?

The aim of this observational cohort study with a control group is to compare consonant perception skills in quiet and in noise in children with typical language and learning development and in children with dyslexia, with and without Speech Sound Disorder (SSD). Three groups were included: A control group of twenty children with normal reading abilities and typical language development, twelve children with dyslexia and typical language development and thirteen children with dyslexia and SSD. All subjects received a consonant recognition test in three different listening conditions (quiet, + 10 and 0 Signal-to-Noise Ratio). In all test conditions, children with dyslexia and SSD had significantly lower consonant recognition scores than the control group and the children with dyslexia and typical language development (p< .0001). The poorer performances observed in children with dyslexia and SSD may be explained by impaired phonological processing underlying both conditions.

Association Between Tinnitus Pitch and Consonant Recognition in Noise.

Purpose Difficulties in speech-in-noise understanding are often reported in individuals with tinnitus. Building on our previous findings that speech-in-noise performance is correlated with subjective loudness of tinnitus, this study aimed to investigate the effect of tinnitus pitch on consonant recognition in noise. Method Pure-tone audiometry and the Quick Speech-in-Noise Test were conducted on 66 participants categorized into four groups by their hearing sensitivity and self-report of tinnitus. Consonant recognition scores at various frequency ranges were obtained at the 5 dB SNR condition of the Quick Speech-in-Noise Test. Participants with tinnitus also completed a tinnitus pitch-matching procedure. Correlation analyses were conducted between tinnitus pitch and the frequency of the worst consonant recognition, and the error rates based on word and sentence position were compared. Results Regardless of hearing sensitivity, tinnitus pitch did not correlate with the frequency of the worst consonant recognition. Sentence-initial word recognition was affected by hearing loss, whereas sentence-final word recognition was not affected by hearing loss or tinnitus. In contrast to individuals with normal hearing, participants with hearing loss varied in full-sentence recognition, with those reporting tinnitus exhibiting significantly higher error rates. Conclusions The findings suggest that the effect of tinnitus on consonant recognition in noise may involve higher level functions more than perceptual characteristics of tinnitus. Furthermore, for individuals with speech-in-noise concerns, clinical evaluation should address both hearing sensitivity and the presence of tinnitus. Future speech-in-noise studies should incorporate cognitive tests and, possibly, brain imaging to parse out the contribution of cognitive factors, such as cognitive control, in speech-in-noise in tinnitus.

Predicting consonant recognition and confusions in normal-hearing listeners.

The perception of consonants in background noise has been investigated in various studies and was shown to critically depend on fine details in the stimuli. In this study, a microscopic speech perception model is proposed that represents an extension of the auditory signal processing model by Dau, Kollmeier, and Kohlrausch [(1997). J. Acoust. Soc. Am. 102, 2892-2905]. The model was evaluated based on the extensive consonant perception data set provided by Zaar and Dau [(2015). J. Acoust. Soc. Am. 138, 1253-1267], which was obtained with normal-hearing listeners using 15 consonant-vowel combinations mixed with white noise. Accurate predictions of the consonant recognition scores were obtained across a large range of signal-to-noise ratios. Furthermore, the model yielded convincing predictions of the consonant confusion scores, such that the predicted errors were clustered in perceptually plausible confusion groups. The large predictive power of the proposed model suggests that adaptive processes in the auditory preprocessing in combination with a cross-correlation based template-matching back end can account for some of the processes underlying consonant perception in normal-hearing listeners. The proposed model may provide a valuable framework, e.g., for investigating the effects of hearing impairment and hearing-aid signal processing on phoneme recognition.

Enhancing speech envelope by integrating hair-cell adaptation into cochlear implant processing

Cochlear implants (CIs) bypass some of the mechanisms that underlie normal neural behavior as occurs in acoustic hearing. One such neural mechanism is short-term adaptation, which has been proposed to have a significant role in speech perception. Acoustically-evoked neural adaptation has been mainly attributed to the depletion of neurotransmitter in the hair-cell to auditory-nerve synapse and is therefore not fully present in CI stimulation. This study evaluated a signal processing method that integrated a physiological model of hair-cell adaptation into CI speech processing. The linear high-pass adaptation process expanded the range of rapid variations of the electrical signal generated by the clinical processing strategy. Speech perception performance with the adaptation-based processing was compared to that of the clinical strategy in seven CI users. While there was large variability across subjects, the new processing improved sentence recognition and consonant identification scores in quiet in all the tested subjects with an average improvement of 8% and 6% respectively. Consonant recognition scores in babble noise were improved at the higher signal-to-noise ratios tested (10 and 6 dB) only. Information transfer analysis of consonant features showed significant improvements for manner and place of articulation features, but not for voicing. Enhancement of within-channel envelope cues was confirmed by consonant recognition results obtained with single-channel strategies that presented the overall amplitude envelope of the signal on a single active electrode. Adaptation-inspired envelope enhancement techniques can potentially improve perception of important speech features by CI users.

Phoneme recognition in vocoded maskers by normal-hearing and aided hearing-impaired listeners.

This study compared modulation benefit for phoneme recognition obtained by normal-hearing (NH) and aided hearing-impaired (HI) listeners. Consonant and vowel recognition scores were measured using nonsense syllables in the presence of a steady-state noise and four vocoded speech maskers. Vocoded maskers were generated by modulating the steady-state noise, in either one or six frequency channels, with the speech envelope extracted from the speech of either a single talker or a four-talker babble. Aided HI listeners obtained lower consonant recognition scores than NH listeners in all masker conditions. Vowel recognition scores for aided HI listeners were comparable to NH scores, except in the six-channel vocoded masker conditions where they were relatively lower. Analysis using the extended speech intelligibility index developed by Rhebergen, Versfeld, and Dreschler [(2006). J. Acoust. Soc. Am. 120(6), 3988-3997] suggested that the signal-to-noise ratio deficit observed in aided HI listeners was largely due to uncompensated audibility loss. There was no significant difference between modulation masking release obtained by NH and aided HI listeners for both consonant and vowel recognition.

Prediction of consonant recognition in quiet for listeners with normal and impaired hearing using an auditory model

Consonant recognition was assessed in normal-hearing (NH) and hearing-impaired (HI) listeners in quiet as a function of speech level using a nonsense logatome test. Average recognition scores were analyzed and compared to recognition scores of a speech recognition model. In contrast to commonly used spectral speech recognition models operating on long-term spectra, a "microscopic" model operating in the time domain was used. Variations of the model (accounting for hearing impairment) and different model parameters (reflecting cochlear compression) were tested. Using these model variations this study examined whether speech recognition performance in quiet is affected by changes in cochlear compression, namely, a linearization, which is often observed in HI listeners. Consonant recognition scores for HI listeners were poorer than for NH listeners. The model accurately predicted the speech reception thresholds of the NH and most HI listeners. A partial linearization of the cochlear compression in the auditory model, while keeping audibility constant, produced higher recognition scores and improved the prediction accuracy. However, including listener-specific information about the exact form of the cochlear compression did not improve the prediction further.

Speech Recognition in Natural Background Noise

In the real world, human speech recognition nearly always involves listening in background noise. The impact of such noise on speech signals and on intelligibility performance increases with the separation of the listener from the speaker. The present behavioral experiment provides an overview of the effects of such acoustic disturbances on speech perception in conditions approaching ecologically valid contexts. We analysed the intelligibility loss in spoken word lists with increasing listener-to-speaker distance in a typical low-level natural background noise. The noise was combined with the simple spherical amplitude attenuation due to distance, basically changing the signal-to-noise ratio (SNR). Therefore, our study draws attention to some of the most basic environmental constraints that have pervaded spoken communication throughout human history. We evaluated the ability of native French participants to recognize French monosyllabic words (spoken at 65.3 dB(A), reference at 1 meter) at distances between 11 to 33 meters, which corresponded to the SNRs most revealing of the progressive effect of the selected natural noise (−8.8 dB to −18.4 dB). Our results showed that in such conditions, identity of vowels is mostly preserved, with the striking peculiarity of the absence of confusion in vowels. The results also confirmed the functional role of consonants during lexical identification. The extensive analysis of recognition scores, confusion patterns and associated acoustic cues revealed that sonorant, sibilant and burst properties were the most important parameters influencing phoneme recognition. . Altogether these analyses allowed us to extract a resistance scale from consonant recognition scores. We also identified specific perceptual consonant confusion groups depending of the place in the words (onset vs. coda). Finally our data suggested that listeners may access some acoustic cues of the CV transition, opening interesting perspectives for future studies.

Mandarin Consonant Contrast Recognition Among Children With Cochlear Implants or Hearing Aids and Normal-Hearing Children

The purpose of the present study was to investigate the consonant recognition of Mandarin-speaking children with cochlear implants (CIs) and hearing aids (HAs) and to determine if they reach a level of consonant recognition similar to that of normal-hearing (NH) children. Little information is available in the literature regarding the consonant perception abilities of prelingually deafened young children with either CIs or HAs. No studies have compared Mandarin-Chinese consonant contrast recognition in CI and HA children. Forty-one prelingually deafened children with CIs, 26 prelingually deafened children with HAs, and 30 NH children participated in this study. The 3 groups were matched for chronologic age (3-5 yr). The hearing-impaired groups were matched for age at fitting of the devices, duration of device use, and aided hearing threshold. All subjects completed a computerized Mandarin consonant phonetic contrast perception test. CI and HA children scored, on average, approximately 8 percentage points below the mean NH group performance on the consonant contrast recognition. Approximately 40% of the CI and HA children had not reached a performance level of the NH group. No significant differences in the consonant recognition scores were found between the CI and HA groups. Age of implantation was correlated with consonant contrast recognition in the CI group. When age at fitting of the devices, duration of device use, and aided thresholds are matched at the group level, consonant recognition is similar between the CI and HA children after 2 years of device use. Early implantation tends to yield better consonant contrast recognition in the young children with CIs. However, a large amount of variance in performance was not accounted for by the demographic variables studied.

The Consonant-Weighted Envelope Difference Index (cEDI): A Proposed Technique for Quantifying Envelope Distortion

The benefits of amplitude compression in hearing aids may be limited by distortion resulting from rapid gain adjustment. To evaluate this, it is convenient to quantify distortion by using a metric that is sensitive to the changes in the processed signal that decrease consonant recognition, such as the Envelope Difference Index (EDI; Fortune, Woodruff, & Preves, 1994). However, the EDI relies on the entire duration of the signal, including portions irrelevant to consonant recognition. This note describes a computationally efficient method of automatically segmenting speech in time according to the syllable structure. Our technique uses the 1st derivative of the envelope as a basis. Peaks located in the derivative were used to generate a weighting function for the computation of a metric of signal distortion. The weighting function significantly improved the variance explained in consonant recognition scores over previous methods. However, only 3.2% of the variance was explained in the revised model. This technique was effective in focusing the analysis of distortion on specific segments of the signal. Use of the technique has implications for speech analysis. The difference in the amplitude envelope of consonants is not a robust model of the effect of hearing aid compression on consonant recognition.

Relative contribution of off- and on-frequency spectral components of background noise to the masking of unprocessed and vocoded speech

The present study examined the relative influence of the off- and on-frequency spectral components of modulated and unmodulated maskers on consonant recognition. Stimuli were divided into 30 contiguous equivalent rectangular bandwidths. The temporal fine structure (TFS) in each "target" band was either left intact or replaced with tones using vocoder processing. Recognition scores for 10, 15 and 20 target bands randomly located in frequency were obtained in quiet and in the presence of all 30 masker bands, only the off-frequency masker bands, or only the on-frequency masker bands. The amount of masking produced by the on-frequency bands was generally comparable to that produced by the broadband masker. However, the difference between these two conditions was often significant, indicating an influence of the off-frequency masker bands, likely through modulation interference or spectral restoration. Although vocoder processing systematically lead to poorer consonant recognition scores, the deficit observed in noise could often be attributed to that observed in quiet. These data indicate that (i) speech recognition is affected by the off-frequency components of the background and (ii) the nature of the target TFS does not systematically affect speech recognition in noise, especially when energetic masking and/or the number of target bands is limited.

The effects of frequency-place shift on consonant confusion in cochlear implant simulations.

The effects of frequency-place shift on consonant recognition and confusion matrices were examined. Frequency-place shift was manipulated using a noise-excited vocoder with 4 to 16 channels. In the vocoder processing, the location of the most apical carrier band varied from the matched condition (i.e., 28 mm from the base of the cochlear) to a basal shift (i.e., 22 mm from the base) in a step size of 1 mm. Ten normal-hearing subjects participated in the 20-alternative forced-choice test, where the consonants were presented in a /Ca/ context. Shift of 3 mm or more caused the consonant recognition scores to decrease significantly. The effects of spectral resolution disappeared when the amount of shift reached >or=3 mm. Information transmitted for voicing and place of articulation varied with spectral shift and spectral resolution, while information transmitted for manner was affected only by spectral shift but not spectral resolution. Spectral shift has shown specific effects on the confusion patterns of the consonants. The direction of errors reversed as spectral shift increased and the patterns of reversal were consistent across channel conditions. Overall, transmission of the consonant features can be accounted for by the acoustic features of the speech signal.

Children’s Speech Recognition Scores: The Speech Intelligibility Index and Proficiency Factors for Age and Hearing Level

The objective of this study was to predict consonant recognition scores of adults, children, and children with hearing impairment, using the Speech Intelligibility Index (SII). It was hypothesized that an adult-derived transfer function would be insufficient to predict the scores for children, and that transfer functions for normally hearing listeners would be insufficient to predict scores for children with hearing impairment. Proficiency corrections for age and hearing loss were explored. A 21-consonant test of speech recognition was applied across five signal to noise ratios in a forced choice procedure. Four adults (aged 27-32 yrs), 15 children with normal hearing (aged 6.6-16.9 yrs), and 14 children with mild to severe hearing loss (aged 7.5-18 yrs) participated. The SII was computed for each listener and each test condition using the one-third octave band method. Transfer functions were fitted to the data of each group. The adult-derived transfer function over-predicted the children's scores. Significant increases in prediction accuracy were obtained when the effects of age and hearing loss were incorporated into the transfer function as proficiency factors. The SII could successfully be used to predict speech recognition scores for both adults and children, once the effects of age and hearing loss were included in the development of a transfer function. Specific proficiency factors developed here may not generalize to other data sets. Nonetheless, the results shed light on factors to consider when using the SII to predict children's speech recognition scores.

Effects of Vowel Context on the Recognition of Initial and Medial Consonants by Cochlear Implant Users

Scores on consonant-recognition tests are widely used as an index of speech-perception ability in cochlear implant (CI) users. The consonant stimuli in these tests are typically presented in the /alpha/ vowel context, even though consonants in conversational speech occur in many other contexts. For this reason, it would be useful to know whether vowel context has any systematic effect on consonant recognition in this population. The purpose of the present study was to compare consonant recognition for the /alpha, i/, and /u/ vowel contexts for consonants presented in both initial (Cv) and medial (vCv) positions. Twenty adult CI users with one of three different implanted devices underwent consonant-confusion testing. Twelve stimulus conditions that differed according to vowel context (/alpha, i, u/), consonant position (Cv, vCv), and talker gender (male, female) were assessed in each subject. Mean percent-correct consonant-recognition scores were slightly (5 to 8%) higher for the /alpha/ and /u/ vowel contexts than for the /i/ vowel context for both initial and medial consonants. This general pattern was observed for both male and female talkers, for subjects with better and poorer average consonant-recognition performance, and for subjects using low, medium, and high stimulation rates in their speech processors. In contrast to the mean data, many individual subjects demonstrated large effects of vowel context. For 10 of 20 subjects, consonant-recognition scores varied by 15% or more across vowel contexts in one or more stimulus conditions. Similar to the mean data, these differences generally reflected better performance for the /alpha/ and /u/ vowel contexts than for the /i/ vowel context. An analysis of consonant features showed that overall performance was best for the voicing feature, followed by the manner and place features, and that the place feature showed the strongest effect of vowel context. Vowel-context effects were strongest for the six consonants /d, j, n, k, m/, and /l/. For three of these consonants (/j, n, k/), the back vowels /alpha/ and /u/ produced substantially (30 to 35%) higher mean scores than the front vowel /i/. For each of the remaining three consonants, a unique pattern was observed in which a different single vowel produced substantially higher scores than the others. Several additional consonants (/s, g, w, b/, and /d/) showed strong context effects in either the initial consonant or medial consonant position. Overall, voiceless stop, nasal, and glide-liquid consonants showed the strongest effects of vowel context, whereas the voiceless fricative and voiceless affricate consonants were least affected. Consistent with the feature analysis, a qualitative assessment of phoneme errors for the six key consonants indicated that vowel-context effects stem primarily from changes in the number of place-of-articulation errors made in each context. Vowel context has small but significant effects on consonant-recognition scores for the "average" CI listener, with the back vowels /alpha/ and /u/ producing better performance than the front vowel /i/. In contrast to the average results, however, the effects of vowel context are sizable in some individual subjects. This suggests that it may be beneficial to assess consonant recognition using two vowels, such as /alpha/ and /i/, which produce better and poorer performance, respectively. The present results underscore previous findings that poor transmission of spectral speech cues limits consonant-recognition performance in CI users. Spectral cue transmission may be hindered not only by poor spectral resolution in these listeners but also by the brief duration and dynamic nature of consonant place-of-articulation cues.

Auditory temporal resolution and speech performance in cochlear implant users

It has been widely assumed that speech recognition depends, among other factors, on the auditory temporal resolution function. The present study was conducted to investigate the relationship between auditory temporal resolution and speech performance in nine Nucleus-22 cochlear implant (CI) listeners fitting with the SPEAK strategy. A modulation detection threshold task was used to evaluate auditory temporal resolution. Modulation detection thresholds were measured as a function of current levels. Results showed some marked difference in the temporal modulation function across CI users. Some patients showed increased modulation thresholds across the entire dynamic range while others showed thresholds asymptoted as low as 30% of the dynamic range. There is a strong correlation between consonant recognition scores and the mean modulation detection thresholds (across the entire dynamic range) among CI users. The correlation between the mean modulation detection thresholds and vowel recognition scores was relatively weak but was still significant. The results suggest that phoneme recognition, especially consonant recognition, depends highly on the patients’ capability to detect temporal modulation. One implication from the present results is that the high variability of speech performance is at least partly due to the differences in the temporal resolution of CI users. [Work supported by NIDCD.]

Co-modulation masking release (CMR) in consonant recognition

In the psychoacoustic phenomenon of co-modulation masking release (CMR), detection of a target in a narrow-band noise is improved in the presence of a co-modulated flanking noise band. The present study investigates whether CMR could improve speech perception. Consonant recognition scores of bandlimited speech, 1200–2200 Hz, were measured under a modulated noise of the same bandwidth, and a second band of noise (5000–7000 Hz), which was either co-modulated or non-co-modulated with the first band. With a hypothesis that CMR improves discrimination of phonemes that are characterized by temporal cues, i.e., amplitude envelope, rather than spectral cues, two groups of consonants were selected for testing: stop consonants (similar temporal envelopes, but different spectra) and alveolar–dental consonants (similar spectra, but different temporal envelopes). Results agreed with the hypothesis: significantly higher recognition scores were obtained with the co-modulated noise band for alveolar–dental consonants, while the difference was marginal for stop consonants. This indicates that CMR can improve speech perception as well as psychophysical thresholds. Co-modulation in the flanking band seems to enable listeners to better ‘‘hear out’’ subtle temporal speech cues in noise, leading to better recognition. [Work supported by Quota scholarship and NIDCD.]

Effects of electrode location and spacing on phoneme recognition with the Nucleus-22 cochlear implant.

The objective of this paper was to determine how phoneme identification was affected by the cochlear location and spacing of the electrodes in cochlear implant listeners. Subjects were initially programmed with the full complement of 20 active electrodes, in which each electrode was assigned to represent the output of one filter in the normal SPEAK processor. In the present study several four-electrode processors were constructed by assigning the output of more than one filter to a single electrode. In all conditions speech sounds were still analyzed into 20 frequency bands and processed according to the usual SPEAK processing strategy, but the location and spacing of the four stimulated electrode pairs were varied systematically. In Experiment I, the spacing between stimulated electrodes was fixed at 3.75 mm and the cochlear location of the four electrode pairs was shifted from the most-apical position up to 3.0 mm toward the base in 0.75 mm steps. In Experiment II, the spatial separation between the four electrode pairs (each bipolar-plus-one) was systematically changed from 1.5 mm to 4.5 mm while holding the most apical active electrode fixed. In Experiment III, the spacing of active electrodes was varied to represent equal tonotopic spacing to equal linear frequency intervals between pairs. Recognition of medial vowels and consonants was measured in three subjects with these custom four-electrode speech processors. In Experiment I, results showed that both vowel and consonant recognition were best when the electrodes were in the most apical locations. In Experiment II, best speech recognition occurred when electrode pairs were separated by 3 to 3.75 mm. In Experiment III, both vowel and consonant recognition scores decreased when the spacing of electrode pairs was changed from equal tonotopic spacing to equal linear frequency intervals. Overall, vowel and consonant recognition were best at the most apical electrode locations and when the spacing of electrodes matched the frequency intervals of the analysis filters. Consonant recognition was relatively robust to alterations in electrode location and spacing. The best vowel scores with four-electrode speech processors were about 10 percentage lower than scores obtained with the full 20-electrode speech processors. However, the best consonant scores with four-electrode speech processors were similar to those obtained with the full 20-electrode speech processors. Information transmission analysis revealed that temporal envelope cues (voicing and manner) were not strongly affected by changes in electrode location and spacing, whereas spectral cues, as represented by vowel recognition and consonantal place of articulation, were strongly affected. Both spectral and temporal phoneme cues were strongly affected by the degree of tonotopic warping, created by altering both the location and spacing of the activated electrodes. The cochlear location and spacing of the activated electrodes had a clear effect on phoneme recognition. Temporal cues were less affected by tonotopic shifts or linear tonotopic stretching or shrinking, but were susceptible to nonlinear tonotopic warping. Spectral cues were sensitive to all tonotopic manipulations: shifting, linear stretching, and nonlinear warping. However, the present experiments could not differentiate whether the optimal mapping between analysis frequency bands and stimulation electrodes was determined by the normal acoustic tonotopic pattern or by the pattern learned from experience with the 20-electrode implant.

The effect of masking and filtering of envelope modulations on the recognition of consonants

Consonant identification scores were obtained for 18 Dutch consonant phonemes in VCV context, in two experiments in which the narrow-band envelopes of the speech stimuli were manipulated by masking or filtering. For this purpose, the speech stimuli were split up into 26 frequency bands, each one critical band wide. The amplitude envelope of each band, as obtained with a Hilbert transform, was disturbed in two ways: (1) by a narrow-band envelope masker around 4, 8, and 12 Hz, and (2) by notch-band filtering around the same frequencies. Next the speech stimuli were synthesized by using narrow-band noise carriers, each a critical band wide, which were modulated by the aforementioned envelopes. Ten normal-hearing subjects participated in the experiments. The confusion matrices were analyzed by means of multidimensional scaling techniques and compared between the two manipulation methods. The results of the filtering experiment show that consonant recognition scores behave like a low-pass function of notch-width when expressed in octaves. The differences with the results of modulation masking will be discussed, especially with respect to modulation transfer function and consonant recognition cues.

The effects of the attenuation of second and third formant frequencies on the recognition of stop consonant vowel syllables in aphasic and nonaphasic subjects.

The effects of the attenuation of second and third formant frequencies on the recognition of stop consonant vowel syllables (/p,b,t,d,k,g/ paired with the vowel /a/) in fluent and nonfluent aphasic and nonaphasic normal subjects were examined. Statistical analyses revealed significant differences between groups, between formant conditions, and between stop consonants. All groups responded similarly under the formant conditions, but the attenuation of F2 appeared to account for the most striking disparity between consonant recognition scores irrespective of group. A series of confusion matrices showed that confusions made by aphasic subjects involved both place of articulation and voicing errors, particularly for nonfluent aphasics, and involved the substitution of more anterior consonants. Important cues for the voicing of anterior consonants appeared to be carried by F2. The finding that more intact speech production abilities appeared to facilitate more accurate speech recognition was interpreted as support for the motor theory of speech perception.