Impact of text-to-speech features on the reading comprehension of children with reading and language difficulties.

This study investigated the reading comprehension scores of students with reading and language difficulties after reading a passage with and without text-to-speech (TTS). Students, ages 8 to 12 years, read five passages under the following conditions: (a) silent read, (b) read aloud, (c) listen only, (d) TTS with no highlighting, and (e) TTS with highlighting. Students answered multiple-choice comprehension questions following each condition. Mixed ANOVAs were performed to determine whether TTS improved reading comprehension. TTS significantly improved comprehension in comparison to no TTS, and specifically, TTS with no highlighting and TTS with highlighting resulted in significantly higher comprehension scores compared to silent read. No other significant differences were found across conditions including between the presentational features of TTS, specifically TTS with no highlighting and TTS with highlighting conditions. Students were grouped as dyslexia only or reading and language impairment based on their test results. Findings suggested that students with dyslexia only scored significantly higher on reading comprehension questions in all reading conditions and derived significantly more benefit in reading comprehension from TTS and the listen only condition compared to students with Reading and Language Impairment. Overall, TTS may be a helpful tool for supporting the reading comprehension of students with reading and language difficulties, particularly for students with dyslexia only; however, further studies are needed to explore the benefits of TTS' presentational features such as highlighting with students with reading and language difficulties.

Listening for Dysphagia: Voice Quality Sequelae of Material in the Airway.

PURPOSE: Postswallow voice abnormality is often assumed to indicate the presence of material in the laryngeal airway (MIA), but prior research has not shown definitive evidence of a causal relationship. This study investigated if endoscopically confirmed MIA (prandial material or secretions) generates specific voice quality attributes that can be perceptually identified by experienced listeners. METHOD: Forty-four dysphagic adults underwent endoscopic evaluation of swallowing. Time-linked audiovisual recordings of pre- and postswallow phonation were analyzed to determine the presence, amount, and location of prandial material or secretions in the larynx during phonation. Expert listeners completed auditory-perceptual ratings of phonation samples. RESULTS: Voice quality ratings did not differ significantly when there was MIA during postswallow phonation. However, judgments of voice quality abnormality for MIA samples were elevated when analyses controlled for effects of baseline dysphonia. Listeners were most likely to perceive voice quality abnormality when larger amounts of secretions were present, as compared to prandial material. Interrater reliability was variable and ranged from low to moderate across perceptual parameters. CONCLUSIONS: MIA during phonation occurred for many participants, but perception of voice quality abnormality was variable when MIA was present. Baseline dysphonia is common among individuals with dysphagia and may limit perception of voice quality abnormalities generated by MIA. Secretions in the laryngeal airway may have a greater impact on voice quality parameters.

Assessment of Intelligibility in Children with Velopharyngeal Insufficiency: The Relationship between Intelligibility in Context Scale and Experimental Measures.

PURPOSE: The purpose of this study was to investigate the clinical application of the Intelligibility in Context Scale (ICS) instrument in children with velopharyngeal insufficiency (VPI). This study investigated the relationship between clinical speech outcomes and parental reports of speech intelligibility across various communicative partners. METHODS: The ICS was completed by the parents of 20 English-speaking children aged 4-12 years diagnosed with VPI. The parents were asked to rate their children's speech intelligibility across communication partners using a 5-point scale. Clinical metrics obtained using standard clinical transcription on the Picture-Cued SNAP-R Test were: (1) percentage of consonants correct (PCC), (2) percentage of vowels correct (PVC), and (3) percentage of phonemes correct (PPC). Nasalance from nasometer data was included as an indirect measure of nasality. Intelligibility scores obtained from naive listener's transcriptions and speech-language pathologists' (SLP) ratings were compared with the ICS results. RESULT: Greater PCC, PPC, PVC, and transcription-based intelligibility values were significantly associated with higher ICS values, respectively (r[20] = 0.84, 0.82, 0.51, and 0.70, respectively; p < 0.05 in all cases). There was a negative and significant correlation between ICS mean scores and SLP ratings of intelligibility (r = -0.74; p < 0.001). There was no significant correlation between ICS values and nasalance scores (r[20] = -0.28; p = 0.22). CONCLUSION: The high correlations obtained between the ICS with PCC and PPC measures indicate that articulation accuracy has had a great impact on parents' decision-making regarding intelligibility in this population. Significant agreement among ICS scores with naive listener transcriptions and clinical ratings supports use of the ICS in practice.

Improved Sensitivity of Digits-in-Noise Test to High-Frequency Hearing Loss.

OBJECTIVES: Hearing loss is most commonly observed at high frequencies. High-frequency hearing loss (HFHL) precedes and predicts hearing loss at lower frequencies. It was previously shown that an automated, self-administered digits-in-noise (DIN) test can be sensitized for detection of HFHL by low-pass filtering the speech-shaped masking noise at 1.5 kHz. This study was designed to investigate whether sensitivity of the DIN to HFHL can be enhanced further using low-pass noise filters with higher cutoff frequencies. DESIGN: The US-English digits 0 to 9, homogenized for audibility, were binaurally presented in different noise maskers including one broadband and three low-pass (cutoff at 2, 4, and 8 kHz) filtered speech-shaped noises. DIN-speech reception thresholds (SRTs) were obtained from 60 normal hearing (NH), and 40 mildly hearing impaired listeners with bilateral symmetric sensorineural hearing loss. Standard and extended high-frequency audiometric pure-tone averages (PTAs) were compared with the DIN-SRTs. RESULTS: Narrower masking noise bandwidth generally produced better (more sensitive) mean DIN-SRTs. There were strong and significant correlations between SRT and PTA in the hearing impaired group. Lower frequency PTALF 0.5,1, 2, 4 kHz had the highest correlation and the steepest slope with SRTs obtained from the 2-kHz filter. Higher frequency PTAHF 4,8,10,12.5 kHz correlated best with SRTs obtained from 4- and 8-kHz filtered noise. The 4-kHz low-pass filter also had the highest sensitivity (92%) and equally highest (with the 8-kHz filter) specificity (90%) for detecting an average PTAHF of 20 dB or more. CONCLUSIONS: Of the filters used, DIN sensitivity to higher frequency hearing loss was greatest using the 4-kHz low-pass filter. These results suggest that low-pass filtered noise may be usefully substituted for broadband noise to improve earlier detection of HFHL using DIN.

Listener characteristics differentially affect self-reported and physiological measures of effort associated with two challenging listening conditions.

Listeners vary in their ability to understand speech in adverse conditions. Differences in both cognitive and linguistic capacities play a role, but increasing evidence suggests that such factors may contribute differentially depending on the listening challenge. Here, we used multilevel modeling to evaluate contributions of individual differences in age, hearing thresholds, vocabulary, selective attention, working memory capacity, personality traits, and noise sensitivity to variability in measures of comprehension and listening effort in two listening conditions. A total of 35 participants completed a battery of cognitive and linguistic tests as well as a spoken story comprehension task using (1) native-accented English speech masked by speech-shaped noise and (2) nonnative accented English speech without masking. Masker levels were adjusted individually to ensure each participant would show (close to) equivalent word recognition performance across the two conditions. Dependent measures included comprehension tests results, self-rated effort, and electrodermal, cardiovascular, and facial electromyographic measures associated with listening effort. Results showed varied patterns of responsivity across different dependent measures as well as across listening conditions. In particular, results suggested that working memory capacity may play a greater role in the comprehension of nonnative accented speech than noise-masked speech, while hearing acuity and personality may have a stronger influence on physiological responses affected by demands of understanding speech in noise. Furthermore, electrodermal measures may be more strongly affected by affective response to noise-related interference while cardiovascular responses may be more strongly affected by demands on working memory and lexical access.

Development and validation of a digits-in-noise hearing test in Persian.

OBJECTIVE: The prevalence of unrecognised and late-diagnosed hearing loss is higher in low- and middle-income than in high-income countries, due in part to lack of access to hearing services. Because hearing screening is important for early identification of hearing loss, development of an accessible, self-screening test that can detect hearing loss reliably and quickly would provide significant benefits, especially for underserved populations. This study aimed to develop and validate a new version of the digits-in-noise (DIN) test for Persian speaking countries. DESIGN: Recordings of Persian digits 0-9 were binaurally presented in broadband speech-shaped noise. Using fitted speech intelligibility functions, digits were homogenised to achieve equal perceptual difficulty across stimuli. The evaluation was established by reference to existing English DIN tests. STUDY SAMPLE: Thirty Persian speaking young adults with normal hearing thresholds (≤20 dB HL, 0.25-8 kHz). RESULTS: Speech intelligibility functions produced a mean speech reception threshold (SRT) of -7.7 dB, corresponding closely to previously developed DIN tests. There was no significant difference between test and retest SRTs, indicating high reliability of the test. Our findings suggest that language-specific factors need to be considered for cross-language comparison of DIN-SRTs. CONCLUSION: This study introduces a convenient tool for future hearing screening in Persian speaking countries with limited access to audiology services.

Social Communication and Structural Language of Girls With High-Functioning Autism Spectrum Disorder.

Purpose The purpose of this study was to characterize social communication and structural language of school-age girls with high-functioning autism spectrum disorder (HF-ASD) compared to a matched group of girls who are typically developing (TD). Method Participants were 37 girls between 7;5 and 15;2 (years;months)-18 HF-ASD and 19 TD. Children completed the Test of Pragmatic Language-Second Edition (TOPL-2) and Clinical Evaluation of Language Fundamentals-Fifth Edition. Parents completed the Children's Communication Checklist-2 United States Edition (CCC-2) and Receptive and Expressive Communication subdomains of the Vineland Adaptive Behavior Scales-Second Edition. Results In the area of social communication, girls with HF-ASD earned significantly lower scores and were more often classified as having an impairment on the TOPL-2 and the CCC-2. However, 28% and 33% earned average scores on the TOPL-2 and the CCC-2, respectively. In the area of structural language, no significant differences were found between groups on Clinical Evaluation of Language Fundamentals-Fifth Edition indexes. In contrast, girls with HF-ASD earned significantly lower scores and were more often classified as having an impairment on the Vineland Adaptive Behavior Scales-Second Edition. Sixty-one percent and 83% scored below average on the Receptive and Expressive Communication subdomains, respectively. Conclusions It has been argued that girls with HF-ASD, when compared to boys with HF-ASD, may have advantages for social communication and structural language that mask their impairments. However, when compared to girls who are TD, girls with HF-ASD demonstrated impaired social communication and structural language. Clinicians should include and carefully examine multiple sources of information when assessing girls with HF-ASD.

Implementation Strategies for the International Dysphagia Diet Standardisation Initiative (IDDSI), Part I: Quantitative Analysis of IDDSI Performance Among Varied Participants.

Purpose The International Dysphagia Diet Standardisation Initiative (IDDSI) is an international standardized framework for texture-modified diets (TMDs). However, user accuracy in conducting IDDSI testing methods are unknown. The aims of this study were threefold: (a) to describe performance on two tasks (an IDDSI knowledge quiz and a TMD sample classification task), (b) to determine interrater and intrarater agreement for classification task performance, and (c) to determine predictive relationships between socioeconomic factors or prior knowledge on task performance. Method Sixty-eight participants were recruited, including health care professionals and non-health care-related individuals. A mixed between-subjects and within-subject design was used. All participants completed a baseline knowledge quiz, 30 min of self-study using the IDDSI.org curriculum, a post self-study knowledge quiz, and a TMD classification task of 21 TMD samples with representation across all IDDSI levels. Data were collected via electronic survey. Results There was a significant increase (p < .001) between pre and post self-study knowledge quiz scores. On the classification task, unmodified foods and drinks were most accurately classified with thickened liquids (IDDSI Levels 1, 2, 3) most inaccurately classified. At baseline, moderate interrater agreement was found with intrarater agreement ranging from fair to almost perfect among identical samples. No significant predictive relationships were found between classification task performance and socioeconomic status or prior experience. Conclusions Thirty minutes of self-study using the online IDDSI.org curriculum improved baseline IDDSI knowledge check performance. Overall accuracy of TMD classification was low and warrants further evaluation given potential adverse health outcomes secondary to inappropriate TMD presentation. Given no predictive relationships between socioeconomic factors and prior experience on task performance, the IDDSI curriculum and classification task appear accessible to various users, including non-health care-related participants. Future studies should more closely observe testing behavior to further characterize variation in participants' use of the testing methods.

Extended high-frequency hearing enhances speech perception in noise.

Young healthy adults can hear tones up to at least 20 kHz. However, clinical audiometry, by which hearing loss is diagnosed, is limited at high frequencies to 8 kHz. Evidence suggests there is salient information at extended high frequencies (EHFs; 8 to 20 kHz) that may influence speech intelligibility, but whether that information is used in challenging listening conditions remains unknown. Difficulty understanding speech in noisy environments is the most common concern people have about their hearing and usually the first sign of age-related hearing loss. Digits-in-noise (DIN), a widely used test of speech-in-noise perception, can be sensitized for detection of high-frequency hearing loss by low-pass filtering the broadband masking noise. Here, we used standard and EHF audiometry, self-report, and successively higher cutoff frequency filters (2 to 8 kHz) in a DIN test to investigate contributions of higher-frequency hearing to speech-in-noise perception. Three surprising results were found. First, 74 of 116 "normally hearing," mostly younger adults had some hearing loss at frequencies above 8 kHz. Early EHF hearing loss may thus be an easily measured, preventive warning to protect hearing. Second, EHF hearing loss correlated with self-reported difficulty hearing in noise. Finally, even with the broadest filtered noise (≤8 kHz), DIN hearing thresholds were significantly better (P < 0.0001) than those using broadband noise. Sound energy above 8 kHz thus contributes to speech perception in noise. People with "normal hearing" frequently report difficulty hearing in challenging environments. Our results suggest that one contribution to this difficulty is EHF hearing loss.

Modelling category goodness judgments in children with residual sound errors.

This study investigates category goodness judgments of /r/ in adults and children with and without residual speech errors (RSEs) using natural speech stimuli. Thirty adults, 38 children with RSE (ages 7-16) and 35 age-matched typically developing (TD) children provided category goodness judgments on whole words, recorded from 27 child speakers, with /r/ in various phonetic environments. The salient acoustic property of /r/ - the lowered third formant (F3) - was normalized in two ways. A logistic mixed-effect model quantified the relationships between listeners' responses and the third formant frequency, vowel context and clinical group status. Goodness judgments from the adult group showed a statistically significant interaction with the F3 parameter when compared to both child groups (p < 0.001) using both normalization methods. The RSE group did not differ significantly from the TD group in judgments of /r/. All listeners were significantly more likely to judge /r/ as correct in a front-vowel context. Our results suggest that normalized /r/ F3 is a statistically significant predictor of category goodness judgments for both adults and children, but children do not appear to make adult-like judgments. Category goodness judgments do not have a clear relationship with /r/ production abilities in children with RSE. These findings may have implications for clinical activities that include category goodness judgments in natural speech, especially for recorded productions.

Identifiers of Language Impairment for Spanish-English Dual Language Learners.

Purpose The purpose of this study was to determine if a standardized assessment developed for Spanish-English dual language learners (SEDLLs) differentiates SEDLLs with language impairment (LI) from children with typical language better than the translated/adapted Spanish and/or English version of a standardized assessment and to determine if adding informal measure/s to the standardized assessment increases the classification accuracy. Method Standardized and informal language assessment measures were administered to 30 Mexican American 4- to 5-year-old SEDLLs to determine the predictive value of each measure and the group of measures that best identified children with LI and typical language. Discriminant analyses were performed on the data set. Results The Morphosyntax and Semantics subtests of the Bilingual English-Spanish Assessment ( Peña, Gutierrez-Clellen, Iglesias, Golstein, & Bedore, 2014 ) resulted in the largest effect size of the individual assessments with a sensitivity of 93.3% and a specificity of 86.7%. Combining these subtests with mean length of utterance in words from the child's better language sample (English or Spanish) was most accurate in identifying LI and can be used with above 90% confidence. Conclusion The Bilingual English-Spanish Assessment Morphosyntax and Semantics subtests were shown to comprise an effective measure for identifying LI; however, including a language sample is suggested to identify LI with greater accuracy.

Bilingualism leads to greater auditory capacity.

The objective of this article is to investigate the effects of bilingualism on auditory capacity of young adults using a dichotic consonant-vowel (CV) test. Listeners were asked to identify distinct CVs dichotically presented to each ear through headphones. CV identification accuracy in both ears served as a measure of auditory capacity of listeners. Eighty normal hearing participants including 40 bilinguals (23 males and 17 females) and 40 monolinguals (11 males and 29 females) were used as study sample. Members of the bilingual group acquired their second language before entering elementary school. The bilingual listeners had higher mean both-ear-correct scores than did monolingual listeners, indicating a greater auditory capacity in the bilingual group than in the monolingual group. The finding of greater auditory capacity in bilinguals using a task requiring divided attention reflects greater ability to store and recall auditory information in bilinguals. However, the inconsistency of results across studies of bilingual advantages indicates that there is a need for further research in this area using both linguistic and non-linguistic tasks and considering age of acquisition as a possible moderating variable.

Modeling talker- and listener-based sources of variability in babble-induced consonant confusions.

Speech communication often occurs in the presence of noise. Patterns of perceptual errors induced by background noise are influenced by properties of the listener and of the noise and target speech. The present study introduces a modification of multilevel general recognition theory in which talker- and listener-based variability in confusion patterns are modeled as global or dimension-specific scaling of shared, group-level perceptual distributions. Listener-specific perceptual correlations and response bias are also modeled as random variables. This model is applied to identification-confusion data from 11 listeners' identifications of ten tokens of each of four consonant categories-[t], [d], [s], [z]-produced by 20 talkers in CV syllables and masked by 10-talker babble. The results indicate that dimension-specific scaling for both listeners and talkers provides a good account of confusion patterns. These findings are discussed in relation to other recent research showing substantial listener-, talker-, and token-based sources of variability in noise-masked speech perception.

Predicting Intelligibility Deficit in Dysphonic Speech with Cepstral Peak Prominence.

OBJECTIVES: The aim of this study was to examine the potential of cepstral peak prominence (CPP) for predicting the intelligibility deficit in dysphonic speech. METHODS: Sentences from Hearing-in-Noise Test were recorded from 18 speakers with dysphonia and 18 speakers with normal voice. These samples were presented to 60 adults with normal hearing in quiet and noise at signal to noise ratio of +0 dB. Intelligibility was measured by orthographic transcription. Cepstral peak prominence was measured for all samples. Correlation between CPP and intelligibility score was examined. RESULTS: Intelligibility was significantly lower in dysphonic speech than normal speech in the presence of background noise. The correlation between CPP and intelligibility score was moderate when the intelligibility scores were averaged per speaker. CONCLUSIONS: Cepstral peak prominence only moderately predicts intelligibility deficit in dysphonic speech. Accordingly, CPP alone is not sufficient for describing the deficit.

Fidelity of Automatic Speech Processing for Adult and Child Talker Classifications.

Automatic speech processing (ASP) has recently been applied to very large datasets of naturalistically collected, daylong recordings of child speech via an audio recorder worn by young children. The system developed by the LENA Research Foundation analyzes children's speech for research and clinical purposes, with special focus on of identifying and tagging family speech dynamics and the at-home acoustic environment from the auditory perspective of the child. A primary issue for researchers, clinicians, and families using the Language ENvironment Analysis (LENA) system is to what degree the segment labels are valid. This classification study evaluates the performance of the computer ASP output against 23 trained human judges who made about 53,000 judgements of classification of segments tagged by the LENA ASP. Results indicate performance consistent with modern ASP such as those using HMM methods, with acoustic characteristics of fundamental frequency and segment duration most important for both human and machine classifications. Results are likely to be important for interpreting and improving ASP output.

Automated Vocal Analysis of Children With Hearing Loss and Their Typical and Atypical Peers.

OBJECTIVES: This study investigated automatic assessment of vocal development in children with hearing loss compared with children who are typically developing, have language delays, and have autism spectrum disorder. Statistical models are examined for performance in a classification model and to predict age within the four groups of children. DESIGN: The vocal analysis system analyzed 1913 whole-day, naturalistic acoustic recordings from 273 toddlers and preschoolers comprising children who were typically developing, hard of hearing, language delayed, or autistic. RESULTS: Samples from children who were hard of hearing patterned more similarly to those of typically developing children than to the language delayed or autistic samples. The statistical models were able to classify children from the four groups examined and estimate developmental age based on automated vocal analysis. CONCLUSIONS: This work shows a broad similarity between children with hearing loss and typically developing children, although children with hearing loss show some delay in their production of speech. Automatic acoustic analysis can now be used to quantitatively compare vocal development in children with and without speech-related disorders. The work may serve to better distinguish among various developmental disorders and ultimately contribute to improved intervention.

The mapping between phonological categories and acoustic cues in the production of English obstruents.

Phonological feature structure is inherently multidimensional, and decades' worth of research in acoustic phonetics has documented both the complex mappings between features and associated acoustic cues as well as the prosodic modulation of these mappings. Most previous studies have focused on how the mean values of acoustic cues vary in complex ways across multiple phonological dimensions, relying on strong assumptions of statistical independence and/or homogeneity of variance across acoustic measures. The present study probes these assumptions by exploring the mapping between phonological voicing, place, and manner features and 8 acoustic cues from tokens of 14 English consonants produced in onset and coda position. Multivariate linear models exhibiting a variety of feature-cue mappings and between-cue statistical relationships were fit to this corpus of acoustic data. Model comparisons indicate that the best statistical description of the data requires pervasive interactions between features with respect to both the locations and the shapes of phonological categories. The implications of these results for work on the production and perception of phonological contrasts is discussed.

Technical clarification to Silbert and Thomas (2013): "decisional separability, model identification, and statistical inference in the general recognition theory framework".

We offer a minor technical correction to the published proof of part (ii) of the main theorem in Silbert and Thomas (Psychonomic Bulletin & Review, 20, 1-20, 2013) that somewhat limits the scope of the equivalence observed in that work. Specifically, in order for a mean shift integrality with decisional separability to be mimicked by a perceptually separable but nondecisionally separable configuration, one needs to assume stimulus invariance. This holds when all of the covariance matrices in the stimulus configuration are equal to each other. We note that part (i) of the theorem is unaffected by this modification; an empirical finding of perceptual separability and the failure of decisional separability can be mimicked by a perceptually nonseparable, decisionally separable configuration without restricting the covariance matrices to be equal. We also note that stimulus invariance is often assumed in simple designs (e.g., Macmillan & Ornstein in Journal of the Acoustical Society of America, 97, 1261-1285, 1998), due to the implausibility of different perceptual correlations being present within stimuli perched very closely in perceptual space.

Estimating large numbers.

Despite their importance in public discourse, numbers in the range of 1 million to 1 trillion are notoriously difficult to understand. We examine magnitude estimation by adult Americans when placing large numbers on a number line and when qualitatively evaluating descriptions of imaginary geopolitical scenarios. Prior theoretical conceptions predict a log-to-linear shift: People will either place numbers linearly or will place numbers according to a compressive logarithmic or power-shaped function (Barth & Paladino, ; Siegler & Opfer, ). While about half of people did estimate numbers linearly over this range, nearly all the remaining participants placed 1 million approximately halfway between 1 thousand and 1 billion, but placed numbers linearly across each half, as though they believed that the number words "thousand, million, billion, trillion" constitute a uniformly spaced count list. Participants in this group also tended to be optimistic in evaluations of largely ineffective political strategies, relative to linear number-line placers. The results indicate that the surface structure of number words can heavily influence processes for dealing with numbers in this range, and it can amplify the possibility that analogous surface regularities are partially responsible for parallel phenomena in children. In addition, these results have direct implications for lawmakers and scientists hoping to communicate effectively with the public.

Decisional separability, model identification, and statistical inference in the general recognition theory framework.

Recent work in the general recognition theory (GRT) framework indicates that there are serious problems with some of the inferential machinery designed to detect perceptual and decisional interactions in multidimensional identification and categorization (Mack, Richler, Gauthier, & Palmeri, 2011). These problems are more extensive than previously recognized, as we show through new analytic and simulation-based results indicating that failure of decisional separability is not identifiable in the Gaussian GRT model with either of two common response selection models. We also describe previously unnoticed formal implicational relationships between seemingly distinct tests of perceptual and decisional interactions. Augmenting these formal results with further simulations, we show that tests based on marginal signal detection parameters produce unacceptably high rates of incorrect statistical significance. We conclude by discussing the scope of the implications of these results, and we offer a brief sketch of a new set of recommendations for testing relationships between dimensions in perception and response selection in the full-factorial identification paradigm.