Speech Parameters Research Articles

Personalized College English Learning Based on Deep Learning under the Background of Big Data.

Generally, in-depth learning has been extensively employed in numerous industries to enhance the growth of economic globalization since the dawn of the big data age. At the same time, the demand for foreign language talent has risen dramatically, and more and more businesses are steadily raising their English proficiency standards. International interactions, as well as scientific and technical exchanges, are influenced by English communicating competence. As a result, colleges and universities place a high value on students studying English. However, English learning is also a compulsory course in Colleges and universities, which is different from other courses. Therefore, in setting teaching objectives, arranging teaching activities, and compiling teaching contents, we should highlight students' personalized needs and formulate learning plans in combination with the individual requirements of each student. Aiming at this problem, this paper adopts the deep learning method to study the personalized learning of College English. Deep learning algorithm utilization in the development of college students' English learning reduces teachers' workload and focuses more on students' personalized needs, which is beneficial to teachers' teaching and students' learning. By analyzing the deep learning neural network model, the process of English speech recognition is described, and the characteristic parameters of English speech are extracted. Then compare and explain the characteristics of deep learning and shallow learning, analyze the relationship between them, and clarify the route of deep learning. Finally, the effect of College Students' personalized learning is analyzed by random sampling, and the effect of College English personalized learning based on deep learning is explained from two aspects: listening effect and reading practice.

Efficacy of cochlear implants in children with borderline hearing who have already achieved significant language development with hearing aids.

There are still debates about timing and effectiveness of cochlear implants (CI) in pediatric subjects with significant residual hearing who do not belong to traditional indication of CI. In this study, we aimed to investigate the outcomes of CI, specifically on improvement of pronunciation, among hearing-impaired children already with a substantial degree of language skills as evaluated by Categories of Auditory Perception (CAP) scores or sentence score. Our cohort comprised pediatric CI recipients from July 2018 through October 2020. Among them, cases with CAP scores of 5 or 6 preoperatively were defined as "borderline cases". We investigated prevalence and etiologies, and compared speech evaluation data preoperatively and postoperatively at three time points (3, 6 and 9-12 months after implantation). Among 86 pediatric CI recipients, 13 subjects (15.12%) had language development that reached CAP scores of 5 or 6 before implantation. Postoperative speech evaluation data 6 months after implantation revealed significant improvement of pronunciation (Urimal Test of Articulation and Phonation scores: UTAP), Infant-Toddler Meaningful Auditory Integration Scale (IT-MAIS) and word perception scores, but not of CAP and sentence perception scores. Notably, the significant improvement of pronunciation based on UTAP scores outstripped that of other speech parameters and this continued steadily up to one-year postoperatively. The result of the study serves as evidence for what to expect from cochlear implantation in hearing-impaired children who have already achieved a substantial degree of language development in terms of CAP scores or sentence perception scores, preoperatively.

Synthesis of EEG signals modeled using non-linear oscillator based on speech data with EKF

In the existing literature, one of the major issue in regression technique is the noisy training data. In this research work, the mathematical model for parameter estimation of neurological signals is developed for investigation of Brain dynamics. This proposed model is developed by using non-linear dynamics of oscillator models. It is assumed that state of the brain of a given person is characterized by an unknown parameter vector ‘θ’ that parameterizes a coupled nonlinear oscillator differential equation model which generates a harmonic process serving as a model for the EEG data collected on the brain surface. It is assumed that the EEG data and speech data of the same person are correlated, so that the speech data obeys another differential equation with a parameter ‘ϕ’ that is a fixed function of the EEG a parameter ‘θ’. ‘θ’ and ‘ϕ’ are estimated by an EKF based on training process based on joint the EEG and speech data. This enables us to get accurate estimates of ‘θ’ and ‘ϕ’ for each person. This training process is repeated for ‘N’ persons and accordingly a parameter vector is obtained for each person. This completes the training process. The validation of this model involves choosing a fresh person not belonging to the training set and estimating the EEG parameter vector for this person based only on his speech data. For this, a function ‘ψ’ which correlates the speech parameter ‘ϕ’ with the EEG parameter ‘θ’ as ‘ϕ=ψ(θ̂)’ is estimated by an affine linear relationship which is substituted into the speech model. The EEG model is then used to generate this person’s EEG data and we compare this with the true EEG data of his brain. This validation scheme can also be used to classify the fresh person’s brain by comparing his parameters with those obtained from the training set.

Can AI make people happy? The effect of AI-based chatbot on smile and speech in Parkinson's disease

IntroductionApproaches for objectively measuring facial expressions and speech may enhance clinical and research evaluation in telemedicine, which is widely employed for Parkinson's disease (PD). This study aimed to assess the feasibility and efficacy of using an artificial intelligence-based chatbot to improve smile and speech in PD. Further, we explored the potential predictive value of objective face and speech parameters for motor symptoms, cognition, and mood. MethodsIn this open-label randomized study, we collected a series of face and conversational speech samples from 20 participants with PD in weekly teleconsultation sessions for 5 months. We investigated the effect of daily chatbot conversations on smile and speech features, then we investigated whether smile and speech features could predict motor, cognitive, and mood status. ResultsA repeated-measures analysis of variance revealed that the chatbot conversations had a significant interaction effect on the mean and standard deviation of the smile index during smile sections (both P = .02), maximum duration of the initial rise of the smile index (P = .04), and frequency of filler words (P = .04), but no significant interaction effects were observed for clinical measurements including motor, cognition, depression, and quality of life. Explorative analysis using statistical and machine-learning models revealed that the smile indices and several speech features were associated with motor symptoms, cognition, and mood in PD. ConclusionAn artificial intelligence-based chatbot may positively affect smile and speech in PD. Smile and speech features may capture the motor, cognitive, and mental status of patients with PD.

Comparison of different Classification techniques for the detection of Speech affected due to Respiratory disorders

Respiratory disorders are one of the common chronic disorders across the world. Any malfunctioning in the respiration process affects speech as speech and respiration go hand in hand. Analysis of speech parameter variation due to respiratory disorders is a prime research area in today’s scenario. In this paper, an efficient approach using a machine learning paradigm was used to detect the speech affected due to respiratory disorders. Various speech parameters were extracted (F1, F2, F3… etc.) using PRAAT software along with MFCC and LPC coefficients. Statistically, significant features were obtained to determine the predominant parameters. The extracted features were applied to different machine learning paradigms then different classification techniques were analyzed and compared to distinguish between normal and affected speech.5-fold,10-fold cross-validation, and hold-out data division protocols were used for evaluating the classification results. To evaluate the performance of the proposed methodology, Accuracy, sensitivity, specificity, and the area under receiver operating characteristics (AUC) were used. The result demonstrates that the speech parameters (F1, F2, F3…etc.) extraction method and the LPC method achieved a significantly higher classification accuracy, sensitivity, and specificity of 100% and AUC of 1, under the holdout method. While MFCC method achieved the highest classification accuracy of 83%, sensitivity of 100%, specificity of 67%, and AUC of 0.83.

Training speech pathology students in the perceptual evaluation of speech in patients with cleft palate: Reliability results and the students' perspective

IntroductionListener training is necessary to achieve agreement between perceptual ratings. However, evidence is limited regarding the effect of a training in the perceptual evaluation of cleft palate speech on the reliability of students in speech pathology. The aim of this study was to evaluate the short-term and longer-term effect of a training using the Belgian Dutch outcome tool for cleft palate speech on reliability and the students' experiences. MethodsA 2-h training was provided during a specialist course for 31 first year master students in speech pathology. Information regarding listener characteristics, speech documentation and speech analysis (speech parameters and rating scales) was discussed. This last part focused on the Belgian Dutch outcome tool for speech in patients with cleft palate, including reference samples and consensus listening exercises. Pre training (T1) and post-training reliability (immediately following the training (T2) and one month later (T3)) were collected for the speech variables speech understandability, speech acceptability, hypernasality, hyponasality, nasal emission and nasal turbulence that were rated by means of ordinal scales. Inter-rater reliability was determined with interclass correlation coefficients whereas percentages of agreement were calculated for the intra-rater reliability. ResultsQualitative inter-rater ICC categories improved for hyponasality and speech acceptability at T2 and T3, whereas the inter-rater reliability of the other variables remained stable regarding their interpreted categories. All ICC values quantitatively improved at T2, except hypernasality and nasal emission. At T3, three of those variables showed a small drop one month later. Regarding intra-rater reliability, a significant decrease for hypernasality and nasal turbulence was found at T2, whereas a significant increase was observed for nasal turbulence and speech acceptability at T3. All variables, except hypernasality, had improved intra-rater reliability at T3 in comparison with T1. Students experienced fewer difficulties with most variables after training. Evaluation of nasal emission and nasal turbulence remained however difficult and required, in the students’ opinion, more attention. Discussion & conclusionOverall a positive training effect was noted, especially regarding the inter-rater reliability and the students’ confidence in perceptual ratings. However, it remains a challenge to achieve a positive training effect for all target variables that is maintained over-time.

Temporal Speech Parameters Detect Mild Cognitive Impairment in Different Languages: Validation and Comparison of the Speech-GAP Test® in English and Hungarian.

The development of automatic speech recognition (ASR) technology allows the analysis of temporal (time-based) speech parameters characteristic of mild cognitive impairment (MCI). However, no information has been available on whether the analysis of spontaneous speech can be used with the same efficiency in different language environments. The main goal of this international pilot study is to address the question of whether the Speech-Gap Test® (S-GAP Test®), previously tested in the Hungarian language, is appropriate for and applicable to the recognition of MCI in other languages such as English. After an initial screening of 88 individuals, English-speaking (n = 33) and Hungarianspeaking (n = 33) participants were classified as having MCI or as healthy controls (HC) based on Petersen's criteria. The speech of each participant was recorded via a spontaneous speech task. Fifteen temporal parameters were determined and calculated through ASR. Seven temporal parameters in the English-speaking sample and 5 in the Hungarian-speaking sample showed significant differences between the MCI and the HC groups. Receiver operating characteristics (ROC) analysis clearly distinguished the English-speaking MCI cases from the HC group based on speech tempo and articulation tempo with 100% sensitivity, and on three more temporal parameters with high sensitivity (85.7%). In the Hungarian-speaking sample, the ROC analysis showed similar sensitivity rates (92.3%). The results of this study in different native-speaking populations suggest that changes in acoustic parameters detected by the S-GAP Test® might be present across different languages.

Speech waveform reconstruction from speech parameters for an effective text to speech synthesis system using minimum phase harmonic sinusoidal model for Punjabi

Speech waveform reconstruction from speech parameters for an effective text to speech synthesis system using minimum phase harmonic sinusoidal model for Punjabi

Multivariable prediction models for health care spending using machine learning: a protocol of a systematic review

BackgroundWith rising cost pressures on health care systems, machine-learning (ML)-based algorithms are increasingly used to predict health care costs. Despite their potential advantages, the successful implementation of these methods could be undermined by biases introduced in the design, conduct, or analysis of studies seeking to develop and/or validate ML models. The utility of such models may also be negatively affected by poor reporting of these studies. In this systematic review, we aim to evaluate the reporting quality, methodological characteristics, and risk of bias of ML-based prediction models for individual-level health care spending.MethodsWe will systematically search PubMed and Embase to identify studies developing, updating, or validating ML-based models to predict an individual’s health care spending for any medical condition, over any time period, and in any setting. We will exclude prediction models of aggregate-level health care spending, models used to infer causality, models using radiomics or speech parameters, models of non-clinically validated predictors (e.g., genomics), and cost-effectiveness analyses without predicting individual-level health care spending. We will extract data based on the Checklist for Critical Appraisal and Data Extraction for Systematic Reviews of Prediction Modeling Studies (CHARMS), previously published research, and relevant recommendations. We will assess the adherence of ML-based studies to the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) statement and examine the inclusion of transparency and reproducibility indicators (e.g. statements on data sharing). To assess the risk of bias, we will apply the Prediction model Risk Of Bias Assessment Tool (PROBAST). Findings will be stratified by study design, ML methods used, population characteristics, and medical field.DiscussionOur systematic review will appraise the quality, reporting, and risk of bias of ML-based models for individualized health care cost prediction. This review will provide an overview of the available models and give insights into the strengths and limitations of using ML methods for the prediction of health spending.

Automatic screening of mild cognitive impairment and Alzheimer’s disease by means of posterior-thresholding hesitation representation

Dementia is a chronic or progressive clinical syndrome, characterized by the deterioration of problem-solving skills, memory and language. In Mild Cognitive Impairment (MCI), which is often considered to be the prodromal stage of dementia, there is also a subtle deterioration of these cognitive functions; however, it does not affect the patients’ ability to carry out simple everyday activities. The timely identification of MCI could provide more effective therapeutic interventions to delay progression, and to postpone the possible conversion to dementia. Since language changes in MCI are present even before the manifestation of other distinctive cognitive symptoms, a non-invasive way of early automatic screening could be the use of speech analysis. Earlier, our research team developed a set of temporal speech parameters that mainly focus on the amount of silence and hesitation, and demonstrated its applicability for MCI detection. However, for the automatic extraction of these attributes, the execution of a full Automatic Speech Recognition (ASR) process is necessary. In this study we propose a simpler feature extraction approach, which still quantifies the amount of silence and hesitation in the speech of the subject, but does not require the application of a full ASR system. We experimentally demonstrate that this approach, operating directly on the frame-level output of a HMM/DNN hybrid acoustic model, is capable of extracting attributes as useful as the ASR-based temporal parameter extraction workflow was able to. That is, on our corpus consisting of 25 healthy controls, 25 MCI and 25 mild AD subjects, we achieve a (three-class) classification accuracy of 70.7%, an F-measure score of 89.6 and a mean AUC score of 0.804. We also show that this approach can be applied on simpler, context-independent acoustic states with only a slight degradation of MCI and mild Alzheimer’s detection performance. Lastly, we investigate the usefulness of the three speaker tasks which are present in our recording protocol.

Whispered Speech Conversion Based on the Inversion of Mel Frequency Cepstral Coefficient Features

A conversion method based on the inversion of Mel frequency cepstral coefficient (MFCC) features was proposed to convert whispered speech into normal speech. First, the MFCC features of whispered speech and normal speech were extracted and a matching relation between the MFCC feature parameters of whispered speech and normal speech was developed through the Gaussian mixture model (GMM). Then, the MFCC feature parameters of normal speech corresponding to whispered speech were obtained based on the GMM and, finally, whispered speech was converted into normal speech through the inversion of MFCC features. The experimental results showed that the cepstral distortion (CD) of the normal speech converted by the proposed method was 21% less than that of the normal speech converted by the linear predictive coefficient (LPC) features, the mean opinion score (MOS) was 3.56, and a satisfactory outcome in both intelligibility and sound quality was achieved.

Many Changes in Speech through Aging Are Actually a Consequence of Cognitive Changes.

Background: During aging, changes in human speech may arise because of the neurophysiological deterioration associated with age, or as the result of an impairment in the cognitive processes underlying speech production. Some speech parameters show specific alterations under the presence of dementia. The objective of our study is to identify which of these parameters change because of age, cognitive state, or the interaction of both. Methods: The sample includes 400 people over 55 years old, who were divided into four groups, according to their age. The cognitive state of the participants was assessed through the MMSE test and three ranks were stablished. Gender was also considered in the analysis. Results: Certain temporal, fluency, rhythm, amplitude and voice quality parameters were found to be related to the cognitive state, while disturbance parameters changed due to age. Frequency parameters were exclusively influenced by gender. Conclusions: Understanding how speech parameters are specifically affected by age, cognitive state, or the interaction of both, is determinant to advance in the use of speech as a clinical marker for the detection of cognitive impairments.

Evaluating a Speech-Specific and a Computerized Step-Training-Specific Rhythmic Intervention in Parkinson's Disease: A Cross-Over, Multi-Arms Parallel Study.

Background:Recent studies suggest movements of speech and gait in patients with Parkinson's Disease (PD) are impaired by a common underlying rhythmic dysfunction. If this being the case, motor deficits in speech and gait should equally benefit from rhythmic interventions regardless of whether it is a speech-specific or step-training-specific approach.Objective:In this intervention trial, we studied the effects of two rhythmic interventions on speech and gait. These rhythmic intervention programs are similar in terms of intensity and frequency (i.e., 3x per week, 45 min-long sessions for 4 weeks in total), but differ regarding therapeutic approach (rhythmic speech vs. rhythmic balance-mobility training).Methods:This study is a cross-over, parallel multi-arms, single blind intervention trial, in which PD patients treated with rhythmic speech-language therapy (rSLT; N = 16), rhythmic balance-mobility training (rBMT; N = 10), or no therapy (NT; N = 18) were compared to healthy controls (HC; N = 17; matched by age, sex, and education: p > 0.82). Velocity and cadence in speech and gait were evaluated at baseline (BL), 4 weeks (4W-T1), and 6 months (6M-T2) and correlated.Results:Parameters in speech and gait (i.e., speaking and walking velocity, as well as speech rhythm with gait cadence) were positively correlated across groups (p < 0.01). Statistical analyses involved repeated measures ANOVA across groups and time, as well as independent and one-samples t-tests for within groups analyses. Statistical analyses were amplified using Reliable Change (RC) and Reliable Change Indexes (RCI) to calculate true clinically significant changes due to the treatment on a patient individual level. Rhythmic intervention groups improved across variables and time (total Mean Difference: 3.07 [SD 1.8]; 95% CI 0.2–11.36]) compared to the NT group, whose performance declined significantly at 6 months (p < 0.01). HC outperformed rBMT and NT groups across variables and time (p < 0.001); the rSLT performed similarly to HC at 4 weeks and 6 months in speech rhythm and respiration.Conclusions:Speech and gait deficits in PD may share a common mechanism in the underlying cortical circuits. Further, rSLT was more beneficial to dysrhythmic PD patients than rBMT, likely because of the nature of the rhythmic cue.

Acoustic characteristics of voice and speech in Arabic-speaking stuttering children

BackgroundUsing different methodologies, several researchers have reported certain acoustical and physiological differences between fluent utterances of stutterers and normally fluent speakers. The aim of this study was to determine acoustic characteristics of voice and speech in Arabic-speaking stuttering children in comparison to normal children and correlate these characteristics with stuttering severity. A sample of 80 Arabic-speaking Egyptian children (including 40 typically developing children and 40 stuttering children) in the age range 5–8 years were subjected to acoustic analysis of voice and speech using the Praat software.ResultsThe stuttering children showed significantly higher values of jitter and shimmer in prolonged /a/ vowel sample, as compared to the normal group. This may reflect the subtle differences in laryngeal functioning or in the complex interaction among the laryngeal, respiratory, and the vocal tract systems in stuttering children. Both jitter and shimmer of prolonged /a/ vowel demonstrated significant positive moderate correlation with stuttering severity as assessed by SSI3. F0 was significantly higher in females than in males, both in normal and stuttering children.ConclusionsThe present study revealed significant differences in the acoustic parameters of voice and speech between Arabic-speaking stuttering children and normal children. Some of these acoustic parameters were significantly correlated with stuttering severity. Acoustic analysis can be used as simple, quick, and cheap tool for assessment of stuttering in children and might be a valuable addition to the diagnostic set for assessment of stuttering severity.

Некоторые особенности регистрации потенциала действия слухового нерва у пациентов с кохлеарными имплантами

Recording of auditory nerve action potential is now widely used in clinical practice to objectively adjust speech processor parameters in patients with cochlear implants. According to some authors, the correlation coefficient between the potential of the auditory nerve and the most comfortable level varies within a fairly wide range, but nevertheless it well determines the relief of the tuning curve. This article discusses the relationship of this correlation to the rate of growth of the response amplitude of the auditory nerve potential. Patients with high growth rate of auditory nerve potential amplitude show high correlation (R = 0,71) of this threshold with maximum comfortable level of auditory perception, but patients with low speed show low correlation (R = 0,34) of this dependence. In the latter case, the automatic mode of recording the auditory nerve potential does not achieve the desired result and requires the use of the visual analysis mode of the obtained results. In this case, it is necessary to create several progressive maps with subsequent verification of the results on the patient's behavioral reactions. The paper also presents a variety of options for recording the effect potential of the auditory nerve in special groups of patients with an anomaly of the cochlear and auditory nerve.

Organização temporal da fala em três condições neurológicas relacionadas aos núcleos da base

ABSTRACT Basal ganglia participate in neural networks that control voluntary body movements, including speech. Diseases that affect the function of these structures can generate abnormal hypokinetic or hyperkinetic movements, influencing speech motor control. How does prosodic temporal organization vary in dysarthria due to Parkinson’s disease (PD), Huntington’s disease (HD), and Sydenham’s chorea (SC)? Three clinical groups (PD with and without medication, HD, and SC) of 15 participants and a control group (n = 18) read a text aloud. Speech fluency measures were related to syntactic boundaries within the text. There was no correlation between global motor scales and temporal parameters of speech. There were correlations between syntactic limits and the duration of pauses in all groups. Only the HD results differed from the other clinical and control groups. Clinical groups are slower to produce speech but preserve the syntactic function of prosody at different levels. Basal nuclei dysfunction appears to affect all clinical groups, regardless of etiology.

Analysis of Connected Speech in Persian Aphasic Patients and its Relationship With Type and Severity of Aphasia

Background and Objectives: There are several tests to determine the type and severity of aphasia, but they take a long time to administer when assessing aphasic patients. In recent years, the analysis of spontaneous speech has gained great attention because it is important to diagnose and follow post-treatment improvement in aphasic patients. This study was done to assess some parameters of connected speech in aphasic patients. In addition, the correlation between connected speech parameters and the type and severity of aphasia was measured. Methods: We compared the connected speech parameters of 27 aphasics (10 fluent and 17 non-fluent), Persian speakers, compared with the control group. There were two groups matched by age, sex, and education. Nest’s bird story pictures were used to elicit a speech sample. In the next step, the connected speech was analyzed to define speech parameters, including speech rate, mean length of utterance, number of utterances, total words, content words, function words, nouns, and verbs. Moreover, the severity of aphasia was measured using a Persian Western Aphasia Battery (WAB). Results: The findings showed significant differences between groups in all parameters of the connected speech (P&lt;0.01). The correlation coefficient between speech parameters and severity of aphasia demonstrated that all parameters were highly correlated (r &gt;0.71) with the severity of aphasia (P&lt;0.01) except for speech rate and the number of function words. There were some typical differences between linguistics grammatical and pragmatical characters of different types of aphasia. Conclusion: Connected speech is one of the most sensitive parts of language in all types of aphasic patients. There are some clinical signs for differential diagnosis of aphasia based on speech measures. According to the findings, the type and severity of aphasia and connected speech were highly correlated. Thus, the use of the connected speech analysis is necessary as an assessment tool for the diagnosis of aphasia.

Phonetic Segmentation using a Wavelet-based Speech Cepstral Features and Sparse Representation Classifier

Speech segmentation is the process of dividing speech signal into distinct acoustic blocks that could be words, syllables or phonemes. Phonetic segmentation is about finding the exact boundaries for the different phonemes that composes a specific speech signal. This problem is crucial for many applications, i.e. automatic speech recognition (ASR). In this paper we propose a new model-based text independent phonetic segmentation method based on wavelet packet speech parametrization features and using the sparse representation classifier (SRC). Experiments were performed on two datasets, the first is an English one derived from TIMIT corpus, while the second is an Arabic one derived from the Arabic speech corpus. Results showed that the proposed wavelet packet de composition features outperform the MFCC features in speech segmentation task, in terms of both F1-score and R-measure on both datasets. Results also indicate that the SRC gives higher hit rate than the famous k-Nearest Neighbors (k-NN) classifier on TIMIT dataset.

Classification of sound environment based on subjective response with speech privacy in open plan offices

To implement pleasant open plan offices (OPOs) in the preference aspect based on the subjective responses of the workers, we proposed novel acoustic environment design classification criteria for speech privacy-related parameters. To this end, we created acoustic stimuli through computer simulations and surveyed the related preference responses of test participants. Then, we selected classification criteria based on the survey response results. First, field measurements were performed in two typical OPOs, and a computer simulation model was built. Twelve different sound stimuli were generated by changing the material absorption coefficient conditions for the ceiling and floor inside an office. The investigation of the noise sensitivity and satisfaction of each sound source according to the individuals indicated that the A-weighted sound pressure level at 4 m (Lp,A,S,4m) and spatial decay rate of speech (D2,S) were the major acoustical parameters that determined OPO satisfaction, with Lp,A,S,4m considered first in the OPO design stage. The Lp,A,S,4m class interval was set to 2 dB, and a new four-level classification criteria was proposed, in which the participant response rates of satisfaction were divided according to the change of Lp,A,S,4m into 25% intervals: A (≤ 46 dB), B (46 dB ≤x≤ 48 dB), C (48 dB ≤x≤ 50 dB), and D (≥50 dB). A large difference of two or more satisfaction classes was observed between the high and low noise sensitivity groups for the same Lp,A,S,4m, confirming that the noise sensitivity of workers influences the preferences for OPOs. The proposed classification criteria are expected to be applicable to the design of comfortable OPOs in the future.

Speech Development in Cleft Palate with and without Robin Sequence.

Robin sequence is defined as the triad of micrognathia, glossoptosis, and upper airway obstruction. In up to 85 percent, it is associated with cleft palate. Many studies have reported worse speech development in Robin sequence children after cleft palate repair. The authors investigated speech development in isolated Robin sequence with cleft palate versus children with cleft palate only at the age of 5 to 6 years. All Robin sequence children were treated with the Tübingen palatal plate after birth. Data were collected using the German version of the Great Ormond Street Speech Assessment. Audio and video recordings were reviewed and analyzed separately by two blinded senior phoniatricians based on the German version of the Universal Reporting Parameters for Cleft Palate Speech, and scored to enable comparability of speech outcomes. Forty-four children (Robin sequence, n = 22; cleft palate only, n = 22) were included. Robin sequence children were significantly older at surgery (11.8 months versus 7.1 months; p < 0.001) but younger at study (70.5 months versus 75.2 months; p = 0.035). They also had more severe cleft of the palate (p = 0.006). All children studied showed good to very good speech development without serious impairment. None of the reported parameters on the German version of the Universal Reporting Parameters for Cleft Palate Speech showed significant group differences; the median total score in the Robin sequence group was 23 (interquartile range, 16.5 to 27.5) versus 19 (interquartile range, 17 to 23) in the cleft palate-only group. Statistical analysis revealed no significant effect of group (Z = -1.47; p = 0.14). No group differences in speech development were found at age 5 to 6 years. Isolated Robin sequence does not necessarily represent a risk for impaired speech development. Risk, II.