Improving Dysarthric Speech Segmentation With Emulated and Synthetic Augmentation.

Acoustic features extracted from speech can help with the diagnosis of neurological diseases and monitoring of symptoms over time. Temporal segmentation of audio signals into individual words is an important pre-processing step needed prior to extracting acoustic features. Machine learning techniques could be used to automate speech segmentation via automatic speech recognition (ASR) and sequence to sequence alignment. While state-of-the-art ASR models achieve good performance on healthy speech, their performance significantly drops when evaluated on dysarthric speech. Fine-tuning ASR models on impaired speech can improve performance in dysarthric individuals, but it requires representative clinical data, which is difficult to collect and may raise privacy concerns. This study explores the feasibility of using two augmentation methods to increase ASR performance on dysarthric speech: 1) healthy individuals varying their speaking rate and loudness (as is often used in assessments of pathological speech); 2) synthetic speech with variations in speaking rate and accent (to ensure more diverse vocal representations and fairness). Experimental evaluations showed that fine-tuning a pre-trained ASR model with data from these two sources outperformed a model fine-tuned only on real clinical data and matched the performance of a model fine-tuned on the combination of real clinical data and synthetic speech. When evaluated on held-out acoustic data from 24 individuals with various neurological diseases, the best performing model achieved an average word error rate of 5.7% and a mean correct count accuracy of 94.4%. In segmenting the data into individual words, a mean intersection-over-union of 89.2% was obtained against manual parsing (ground truth). It can be concluded that emulated and synthetic augmentations can significantly reduce the need for real clinical data of dysarthric speech when fine-tuning ASR models and, in turn, for speech segmentation.

[Analysis of language and influencing factors of children with speech disorder in Beijing].

Objective: To investigate the features and influencing factors of language in children with various types of speech disorders. Methods: A case-control study was carried out, 262 children with speech disorder had been diagnosed at the language-speech clinic of the Center of Children's Healthcare, Children's Hospital, Capital Institute of Pediatrics from January 2021 to November 2023, the children with speech sound disorder as the speech sound disorder group, the children with developmental stuttering as the stuttering group. There were 100 typically-developed children who underwent physical checkups at the Center of Healthcare during the same period as the healthy group. All children experienced a standardized evaluation of language with diagnostic receptive and expressive assessment of mandarin-comprehensive(DREAM-C) and questionnaire, One-way ANOVA and LSD test were conducted to compare the differences in overall language, receptive language, expressive language, semantics, and syntax scores among 3 groups of children. According to the results of DREAM-C, the children with speech disorder were divided into language normal group and language delay group. Chi-square test and multivariate Logistic regression were implemented to analyze the association between the linguistic development of children with speech disorder and potential influential factors. Results: There were 145 children in the speech sound disorder group, including 110 males and 35 females respectively, with an age of (5.9±1.0) years; 117 children in the stuttering group, including 91 males and 26 females, with an age of (5.8±1.0) years; 100 children in the healthy group, including 75 males and 25 females, with an age of (5.7±1.2) years. The variations in overall language, expressive language, and syntax scores among 3 groups of children were statistically significant (92±18 vs.96±11 vs. 98±11, 81±18 vs. 84±14 vs. 88±13, 87±16 vs. 89±11 vs. 91±10, F=5.46, 4.69, 3.68, all P<0.05). Pairwise comparison revealed that the speech sound disorder group had lower scores in overall language, expressive language, and syntactic compared to the healthy group, and the differences were statistically significant (all P<0.01) and the overall language score was lower than that of children with stuttering (P<0.05). In terms of overall language and expressive language, there was a statistically significant difference in the incidence of language delay among the three groups of children (15.9% (23/145) vs. 20.5% (24/117) vs. 7.0% (7/100), 46.2% (67/145) vs. 39.3% (46/117) vs. 26.0% (26/100); χ2=7.93, 10.28; both P<0.05). In terms of overall language, the stuttering group took up the highest proportion. In terms of expressive language, the speech sound disorder group accounted for the highest amount. The incidence of language delay in children with speech disorder was 44.3% (116/262). Non-parent-child reading, daily screen time ≥1 hour and screen exposure before 1.5 years of age are risk factors for the development of language in children with speech disorder (OR=1.87, 2.18, 2.01; 95%CI 1.07-3.27, 1.23-3.86, 1.17-3.45; all P<0.01). Negative family history are protective factors for the progress of language ability (OR=0.37, 95%CI 0.17-0.81, P<0.05). Conclusions: Children with speech disorder tend to have easy access to language delay, especially in expressive language and syntax. The occurrence of language delay in children with speech disorder is tightly connected with factors such as the family medical history, parent-child reading, screen time, etc. Attention should be paid to the development of language in children who suffer from speech disorder.

Language-Independent Acoustic Biomarkers for Quantifying Speech Impairment in Huntington's Disease.

PURPOSE: Changes in voice and speech are characteristic symptoms of Huntington's disease (HD). Objective methods for quantifying speech impairment that can be used across languages could facilitate assessment of disease progression and intervention strategies. The aim of this study was to analyze acoustic features to identify language-independent features that could be used to quantify speech dysfunction in English-, Spanish-, and Polish-speaking participants with HD. METHOD: Ninety participants with HD and 83 control participants performed sustained vowel, syllable repetition, and reading passage tasks recorded with previously validated methods using mobile devices. Language-independent features that differed between HD and controls were identified. Principal component analysis (PCA) and unsupervised clustering were applied to the language-independent features of the HD data set to identify subgroups within the HD data. RESULTS: Forty-six language-independent acoustic features that were significantly different between control participants and participants with HD were identified. Following dimensionality reduction using PCA, four speech clusters were identified in the HD data set. Unified Huntington's Disease Rating Scale (UHDRS) total motor score, total functional capacity, and composite UHDRS were significantly different for pairwise comparisons of subgroups. The percentage of HD participants with higher dysarthria score and disease stage also increased across clusters. CONCLUSION: The results support the application of acoustic features to objectively quantify speech impairment and disease severity in HD in multilanguage studies. SUPPLEMENTAL MATERIAL: https://doi.org/10.23641/asha.25447171.

Dynamic Temporal and Tactile Cueing in Young Children With Childhood Apraxia of Speech: A Multiple Single-Case Design.

PURPOSE: Childhood apraxia of speech (CAS) is a multivariate motor speech disorder that requires a motor-based intervention approach. There is limited treatment research on young children with CAS, reflecting a critical gap in the literature given that features of CAS are often in full expression early in development. Dynamic Temporal and Tactile Cueing (DTTC) is a treatment approach designed for children with severe CAS, yet the use of DTTC with children younger than 3 years of age has not been examined. METHOD: A multiple single-case design was employed to examine the use of DTTC in seven children with CAS (aged 2.5-5 years) over the course of 6 weeks of intervention. Changes in word accuracy were measured in treated words from baseline to posttreatment and from baseline to maintenance (6 weeks posttreatment). Generalization of word accuracy changes to matched untreated words was also examined. A linear mixed-effects model was used to estimate the change in word accuracy for treated and untreated words across all children from baseline to posttreatment and to maintenance. A quasi-Poisson regression model was used to estimate mean change and calculate effect sizes for treated and untreated words. RESULTS: Group-level analyses revealed significant changes in word accuracy for treated and untreated words at posttreatment and maintenance. At the child level, six of seven children displayed medium-to-large effect sizes where word accuracy increased in an average of 3.4/5 words across all children. Each child displayed some degree of generalization to untreated targets, specifically for words with the same syllable shape as the treated words. CONCLUSIONS: These results demonstrate that DTTC can yield positive change in some young children with CAS. Key differences in each child's performance are highlighted.

Gerstmann Syndrome Case-Control Study: Correlation between Brain Lesions & Functional Disability.

Deep functional and structural neuroimaging of a series of Gerstmann's syndrome patients required high accuracy, and our results avoided false overlaps of heterogeneous brain lesions by handling each case of our study subjects separately as an individual case regarding functional and neuroimaging tests. Six patients with Gerstmann tetrad (one with dominant acalculia, one with dominant left and right disorientation, two with writing disabilities and two with finger agnosia) and 6 control subjects with close ages were recruited in the current study. In the main phase, we assessed brain activation in response to experimental and interventional settings using neuroimaging techniques (FMRI-Functional Magnetic Resonance Imagingwhere twelve pictures were taken on a Dell inspiration 3T all-body scanner with sequences of echo pictures, 80o angled, TE 35 ms) of the subject's brain to declare lesions existence and locations that might result in one of the four cognitive impairment domains of Gerstman's syndrome tetrad. We assessed statistically significant differences of patient images vs. control images as well as the images of patients presenting specific symptomatic cognitive dysfunction domain vs. the images of patients presenting the three other domains. Neuroimages were analyzed using multiple databases such as T1 weighted and free sequence types. Gerstmann's syndrome is mainly connected to injury in the dominant parietal lobe, so images comparisons and analysis were only restricted to the left parietal lobe region. P values <0.05were only considered as statistically significant difference in comparisons of functional tests time and accuracy of patients vs. in addition to comparisons of brain images parameters of patient group vs. control group and specific symptomatic domain patients vs. other symptomatic domains patients. Regarding functional testing, Patients group took significantly higher time compared to control group. Regarding brain images parameters, patients in each domain showed significantly different lesions compared to other domains. Moreover, control subjects showed no lesions in the left parietal lobe compared to significant lesions in the patient groups. These results oppose the theory of Gerstmann that a common brain structural injury may result in the combination of all of the four symptomatic dysfunction domains. This may be due to the fact that Gerstmann examined incomplete cases which represent a considerable criticism to his scientific basis. Moreover, he excluded patients with speech difficulties and apraxia.

[Features of aphasic disorders in patients with acute cerebrovascular disorder in the posterior cerebral artery basin]. / Osobennosti afaticheskikh rasstroistv u patsientov s ostrym narusheniem mozgovogo krovoobrashcheniya v basseine zadnei mozgovoi arterii.

Aphasia is a systemic disorder of formed speech that develops as a result of local brain lesions. Most aphasias are characterized by damage to secondary cortical fields, which in turn are responsible for the performance of the functions of gnosis and praxis, which explains the variability in the manifestations of speech disorders in patients with acute cerebrovascular accidents. However, it is necessary in each case to diagnose the central pathological mechanism, which underlies the development of the entire syndrome and determines the entire clinical picture. The most important task of a speech therapist-aphasiologist is to qualify the defect, namely to isolate the mechanism and analyze the syndrome in order to select individual methods of corrective restoration. This article presents a case of a patient with an ischemic stroke in the left posterior cerebral artery with the development of amnestic aphasia in combination with alexia without agraphia.

Functional and structural abnormalities of the speech disorders: a multimodal activation likelihood estimation meta-analysis.

Speech disorders are associated with different degrees of functional and structural abnormalities. However, the abnormalities associated with specific disorders, and the common abnormalities shown by all disorders, remain unclear. Herein, a meta-analysis was conducted to integrate the results of 70 studies that compared 1843 speech disorder patients (dysarthria, dysphonia, stuttering, and aphasia) to 1950 healthy controls in terms of brain activity, functional connectivity, gray matter, and white matter fractional anisotropy. The analysis revealed that compared to controls, the dysarthria group showed higher activity in the left superior temporal gyrus and lower activity in the left postcentral gyrus. The dysphonia group had higher activity in the right precentral and postcentral gyrus. The stuttering group had higher activity in the right inferior frontal gyrus and lower activity in the left inferior frontal gyrus. The aphasia group showed lower activity in the bilateral anterior cingulate gyrus and left superior frontal gyrus. Across the four disorders, there were concurrent lower activity, gray matter, and fractional anisotropy in motor and auditory cortices, and stronger connectivity between the default mode network and frontoparietal network. These findings enhance our understanding of the neural basis of speech disorders, potentially aiding clinical diagnosis and intervention.

Echo detection thresholds in big brown bats (Eptesicus fuscus) vary with echo spectral content.

Echolocating big brown bats (Eptesicus fuscus) broadcast downward frequency-modulated sweeps covering the ultrasonic range from 100-23 kHz in two harmonics. They perceive target range from the time delay between each broadcast and its returning echo. Previous experiments indicated that the bat's discrimination acuity for broadcast-echo delay declines when the lowest frequencies (23-35 kHz) in the first harmonic of an echo are removed. This experiment examined whether echo detection is similarly impaired. Results show that detection thresholds for echoes missing these lowest frequencies are raised. Increased thresholds for echoes differing in spectra facilitates the bat's ability to discriminate against clutter.

ClinGen variant curation expert panel recommendations for classification of variants in GAMT, GATM and SLC6A8 for cerebral creatine deficiency syndromes.

Cerebral creatine deficiency syndromes (CCDS) are inherited metabolic phenotypes of creatine synthesis and transport. There are two enzyme deficiencies, guanidinoacetate methyltransferase (GAMT), encoded by GAMT and arginine-glycine amidinotransferase (AGAT), encoded by GATM, which are involved in the synthesis of creatine. After synthesis, creatine is taken up by a sodium-dependent membrane bound creatine transporter (CRTR), encoded by SLC6A8, into all organs. Creatine uptake is very important especially in high energy demanding organs such as the brain, and muscle. To classify the pathogenicity of variants in GAMT, GATM, and SLC6A8, we developed the CCDS Variant Curation Expert Panel (VCEP) in 2018, supported by The Clinical Genome Resource (ClinGen), a National Institutes of Health (NIH)-funded resource. We developed disease-specific variant classification guidelines for GAMT-, GATM-, and SLC6A8-related CCDS, adapted from the American College of Medical Genetics/Association of Molecular Pathology (ACMG/AMP) variant interpretation guidelines. We applied specific variant classification guidelines to 30 pilot variants in each of the three genes that have variants associated with CCDS. Our CCDS VCEP was approved by the ClinGen Sequence Variant Interpretation Working Group (SVI WG) and Clinical Domain Oversight Committee in July 2022. We curated 181 variants including 72 variants in GAMT, 45 variants in GATM, and 64 variants in SLC6A8 and submitted these classifications to ClinVar, a public variant database supported by the National Center for Biotechnology Information. Missense variants were the most common variant type in all three genes. We submitted 32 new variants and reclassified 34 variants with conflicting interpretations. We report specific phenotype (PP4) using a points system based on the urine and plasma guanidinoacetate and creatine levels, brain magnetic resonance spectroscopy (MRS) creatine level, and enzyme activity or creatine uptake in fibroblasts ranging from PP4, PP4_Moderate and PP4_Strong. Our CCDS VCEP is one of the first panels applying disease specific variant classification algorithms for an X-linked disease. The availability of these guidelines and classifications can guide molecular genetics and genomic laboratories and health care providers to assess the molecular diagnosis of individuals with a CCDS phenotype.

Speech Outcomes of Frenectomy for Tongue-Tie Release: A Systematic Review and Meta-Analysis.

OBJECTIVE: Tongue-tie, which is also known as ankyloglossia, is a common condition where the lingual frenulum is unusually tight or short. While most literature investigates the impact of tongue-tie on breastfeeding, recent articles have examined its role in speech production in children. However, these have not previously been reviewed systematically. This study aims to determine the impact of tongue-tie on speech outcomes and assess whether frenectomy can improve speech function. METHODS: In this systematic review, we conducted a comprehensive search of PubMed/MEDLINE, Cochrane Library, Embase, and speechBITE to analyze primary studies investigating the impact of frenectomy for tongue-tie on speech outcomes. We extracted data regarding patient age, male to female ratio, procedure type, follow-up time, and speech outcomes and ran statistical analyses to determine if frenectomy for tongue-tie leads to improvement in speech issues in pediatric patients. Speech outcomes extracted were subjectively measured based on the interpretation of a speech and language pathologist or parent. RESULTS: Our analysis included 10 studies with an average patient age of 4.10 years, and average cohort size of 22.17 patients. Overall, frenectomy for tongue-tie was associated with an improvement in speech articulation (0.78; 95% CI: 0.64-0.87; P < .01). Increasing patient age was found to be negatively correlated with post-frenectomy speech outcomes (P = .01). However, this relationship disappeared in the adjusted model. CONCLUSION: Overall, we conclude that frenectomy is a suitable treatment to correct speech issues in select patients with tongue-tie if caught early in childhood. Despite the limited investigations around speech outcomes post-frenectomy, these results are informative to providers treating tongue-tie.

Exploring Motor Speech Disorders in Low and Minimally Verbal Autistic Individuals: An Auditory-Perceptual Analysis.

PURPOSE: Motor deficits are widely documented among autistic individuals, and speech characteristics consistent with a motor speech disorder have been reported in prior literature. We conducted an auditory-perceptual analysis of speech production skills in low and minimally verbal autistic individuals as a step toward clarifying the nature of speech production impairments in this population and the potential link between oromotor functioning and language development. METHOD: Fifty-four low or minimally verbal autistic individuals aged 4-18 years were video-recorded performing nonspeech oromotor tasks and producing phonemes, syllables, and words in imitation. Three trained speech-language pathologists provided auditory perceptual ratings of 11 speech features reflecting speech subsystem performance and overall speech production ability. The presence, attributes, and severity of signs of oromotor dysfunction were analyzed, as were relative performance on nonspeech and speech tasks and correlations between perceptual speech features and language skills. RESULTS AND CONCLUSIONS: Our findings provide evidence of a motor speech disorder in this population, characterized by perceptual speech features including reduced intelligibility, decreased consonant and vowel precision, and impairments of speech coordination and consistency. Speech deficits were more associated with articulation than with other speech subsystems. Speech production was more impaired than nonspeech oromotor abilities in a subgroup of the sample. Oromotor deficits were significantly associated with expressive and receptive language skills. Findings are interpreted in the context of known characteristics of the pediatric motor speech disorders childhood apraxia of speech and childhood dysarthria. These results, if replicated in future studies, have significant potential to improve the early detection of language impairments, inform the development of speech and language interventions, and aid in the identification of neurobiological mechanisms influencing communication development.

Beyond 'speech delay': Expanding the phenotype of BRPF1-related disorder.

Pathogenic variants in BRPF1 cause intellectual disability, ptosis and facial dysmorphism. Speech and language deficits have been identified as a manifestation of BRPF1-related disorder but have not been systematically characterized. We provide a comprehensive delineation of speech and language abilities in BRPF1-related disorder and expand the phenotype. Speech and language, and health and medical history were assessed in 15 participants (male = 10, median age = 7 years 4 months) with 14 BRPF1 variants. Language disorders were common (11/12), and most had mild to moderate deficits across receptive, expressive, written, and social-pragmatic domains. Speech disorders were frequent (7/9), including phonological delay (6/9) and disorder (3/9), and childhood apraxia of speech (3/9). All those tested for cognitive abilities had a FSIQ ≥70 (4/4). Participants had vision impairment (13/15), fine (8/15) and gross motor delay (10/15) which often resolved in later childhood, infant feeding impairment (8/15), and infant hypotonia (9/15). We have implicated BRPF1-related disorder as causative for speech and language disorder, including childhood apraxia of speech. Adaptive behavior and cognition were strengths when compared to other monogenic neurodevelopmental chromatin-related disorders. The universal involvement of speech and language impairment is noteable, relative to the high degree of phenotypic variability in BRPF1-related disorder.

The Vocal Flutter of Multiple System Atrophy: A Parkinsonian-Type Phenomenon?

BACKGROUND: Early features of multiple system atrophy (MSA) are similar to those in Parkinson's disease (PD), which can challenge differential diagnosis. Identifying clinical markers that help distinguish MSA from forms of parkinsonism is essential to promptly implement the most appropriate management plan. In the context of a thorough neurological evaluation, the presence of a vocal flutter might be considered a potential feature of MSA-parkinsonian type (MSA-P). CASES: This case series describes clinical histories of 3 individuals with MSA-P. In each case, vocal flutter was detected during neurological and motor speech evaluations. It seemed to be a concomitant feature with the constellation of other signs and symptoms that led to the clinical diagnosis. LITERATURE REVIEW: The vocal flutter may be described as pitch and loudness fluctuations during phonation. Different from a vocal tremor, the flutter phenomenon has higher oscillation frequencies. The neuropathological underpinnings of vocal flutter may be related to generalized laryngeal dysfunction that is commonly described in MSA-P. CONCLUSION: Vocal flutter may be a unique speech feature in some individuals who have MSA-P. Future studies using perceptual and acoustic measures of speech are warranted to quantify these observations and directly compare to other MSA variants, PD, and a control group.

Effect of Phonological Awareness-Focused Interventions on Phonological Errors and Phonemic Awareness in Young School-Age Children.

This study aimed to explore the effects of an integrated phonological awareness intervention on phonological errors and phonemic awareness among young school-age children. Three children with at least one phonological error pattern and below-average phonological awareness skills participated in a non-concurrent multiple baseline single-subject design across participants' investigation. The integrated phonological awareness intervention consisted of completing blending and segmenting activities using 20 trained words, with a dose of 70 to 100 productions of the targeted phonological error pattern for 10, 30-minute sessions. All participants showed improvement in the primary dependent variable of percent consonants correct for their targeted error pattern for trained words. Results for percent phonemes correct showed gains for both blending and segmenting for all participants. All the participants transferred targeted skills to untrained words with their error pattern and generalized blending and segmenting to consonant-vowel-consonant words that did not contain their target error pattern in a pretest/posttest. Integrated phonological awareness intervention was an effective method of simultaneously improving speech production and phonemic awareness skills for young school-age children across 5 hours of treatment. The intervention was designed to be replicable by school-based speech-language pathologists seeking to efficiently support students with phonological errors and phonological awareness deficits.

The value of genomic testing in severe childhood speech disorders.

With increasing gene discoveries for severe speech disorders, genomic testing can alter the diagnostic and clinical paradigms, enabling better life outcomes for children and their families. However, evidence on the value of the outcomes generated is lacking, impeding optimal translation into health care. This study aims to estimate the value and uptake of genomic testing for severe childhood speech disorders. A discrete choice experiment was undertaken to elicit preferences for genomic testing from the perspective of the Australian public (n = 951) and parents of children experiencing severe speech disorder (n = 56). Choice attributes associated with genomic testing were identified through focus groups. A Bayesian D-efficient design was used to develop choice scenarios and choice data were analyzed using a panel error component mixed logit model and a latent class model. Statistically significant preferences were identified across all seven attributes. The mean monetary value of the benefits of genomic testing relative to standard diagnostic care in Australia was estimated at AU$7489 (US$5021) and AU$4452 (US$2985) from the perspectives of the Australian public and families with lived experience of severe speech disorders, with a corresponding test uptake of 94.2% and 99.6%. To ensure fair prioritization of genomics, decision-makers need to consider the wide range of risks and benefits associated with genomic information.

Characterizing Speech Errors Across Primary Progressive Apraxia of Speech Subtypes.

PURPOSE: Apraxia of speech (AOS) is a motor speech disorder affecting articulatory planning and speech programming. When AOS is the sole manifestation of neurodegeneration, it is termed primary progressive apraxia of speech (PPAOS). Recent work has shown that there are distinct PPAOS subtypes: phonetic, prosodic, and those that do not clearly align with either (mixed). PPAOS subtypes differ with respect to the predominating motor speech difficulties, as well as disease progression and underlying pathology. Because past studies have determined PPAOS subtype based on clinical impression, the goal of the present study was to quantitatively determine the distribution of speech error types across PPAOS subtypes in a word repetition task and to investigate how word complexity affects the type and number of speech errors across PPAOS subtypes. METHOD: Forty-five patients with PPAOS (13 phonetic, 23 prosodic, and nine mixed) and 45 healthy controls produced multiple repetitions of words that varied in phonetic complexity. Sound additions, deletions, and substitutions/distortions (phonetic errors) and within-word segmentations (prosodic errors) were calculated. RESULTS: All three PPAOS groups produced significantly more errors than controls, but the total number of errors was comparable among subtypes. The phonetic group produced more phonetic-type errors compared to the prosodic group but comparable to the mixed group. The prosodic group produced more segmentations compared to the phonetic and mixed PPAOS groups. As word complexity increased, the total number of errors increased for PPAOS patients. The phonetic and prosodic groups were more likely to produce phonetic- and prosodic-type errors, respectively, as word complexity increased. CONCLUSIONS: This study provides novel quantitative data showing that PPAOS subtype can be supported by the type and distribution of speech errors in a word repetition task. This may facilitate earlier, more reliable differential diagnosis and aid in disease prognosis, as PPAOS subtypes have distinct disease trajectories.

Detecting bulbar amyotrophic lateral sclerosis (ALS) using automatic acoustic analysis.

Automatic speech assessments have the potential to dramatically improve ALS clinical practice and facilitate patient stratification for ALS clinical trials. Acoustic speech analysis has demonstrated the ability to capture a variety of relevant speech motor impairments, but implementation has been hindered by both the nature of lab-based assessments (requiring travel and time for patients) and also by the opacity of some acoustic feature analysis methods. These challenges and others have obscured the ability to distinguish different ALS disease stages/severities. Validation of automated acoustic analysis tools could enable detection of early signs of ALS, and these tools could be deployed to screen and monitor patients without requiring clinic visits. Here, we sought to determine whether acoustic features gathered using an automated assessment app could detect ALS as well as different levels of speech impairment severity resulting from ALS. Speech samples (readings of a standardized, 99-word passage) from 119 ALS patients with varying degrees of disease severity as well as 22 neurologically healthy participants were analyzed, and 53 acoustic features were extracted. Patients were stratified into early and late stages of disease (ALS-early/ALS-E and ALS-late/ALS-L) based on the ALS Functional Ratings Scale-Revised bulbar score (FRS-bulb) (median [interquartile range] of FRS-bulbar scores: 11[3]). The data were analyzed using a sparse Bayesian logistic regression classifier. It was determined that the current relatively small set of acoustic features could distinguish between ALS and controls well (area under receiver-operating characteristic curve/AUROC = 0.85), that the ALS-E patients could be separated well from control participants (AUROC = 0.78), and that ALS-E and ALS-L patients could be reasonably separated (AUROC = 0.70). These results highlight the potential for automated acoustic analyses to detect and stratify ALS.

Relation of Speech-Language Profile and Communication Modality to Participation of Children With Cerebral Palsy.

PURPOSE: This study aimed to examine the contribution of speech motor impairment (SMI), language impairment, and communication modality to communicative and overall participation outcomes in school-age children with cerebral palsy (CP). METHOD: Eighty-one caregivers of children with CP provided information about their child's speech and language skills, communication modality, and participation through a web-based survey. Caregiver responses to two validated scales were used to quantify children's communicative participation and overall participation. Children were classified into four speech-language profile groups and three communication modality groups for comparison, based on caregiver-reported information regarding their child's communication skills. RESULTS: Children with CP who had co-occurring SMI and language impairment had significantly lower levels of communicative participation and involvement in activities overall, compared to children with SMI alone. Among children with SMI, augmentative and alternative communication (AAC) use was associated with greater overall frequency of participation and involvement in life activities. CONCLUSION: Children with CP who have both SMI and language impairment and those who are nonspeaking communicators should be prioritized early for communication interventions focused on maximizing participation, including consideration of AAC.

Diagnosis of pathological speech with streamlined features for long short-term memory learning.

BACKGROUND: Pathological speech diagnosis is crucial for identifying and treating various speech disorders. Accurate diagnosis aids in developing targeted intervention strategies, improving patients' communication abilities, and enhancing their overall quality of life. With the rising incidence of speech-related conditions globally, including oral health, the need for efficient and reliable diagnostic tools has become paramount, emphasizing the significance of advanced research in this field. METHODS: This paper introduces novel features for deep learning in the analysis of short voice signals. It proposes the incorporation of time-space and time-frequency features to accurately discern between two distinct groups: Individuals exhibiting normal vocal patterns and those manifesting pathological voice conditions. These advancements aim to enhance the precision and reliability of diagnostic procedures, paving the way for more targeted treatment approaches. RESULTS: Utilizing a publicly available voice database, this study carried out training and validation using long short-term memory (LSTM) networks learning on the combined features, along with a data balancing strategy. The proposed approach yielded promising performance metrics: 90% accuracy, 93% sensitivity, 87% specificity, 88% precision, an F1 score of 0.90, and an area under the receiver operating characteristic curve of 0.96. The results surpassed those obtained by the networks trained using wavelet-time scattering coefficients, as well as several algorithms trained with alternative feature types. CONCLUSIONS: The incorporation of time-frequency and time-space features extracted from short segments of voice signals for LSTM learning demonstrates significant promise as an AI tool for the diagnosis of speech pathology. The proposed approach has the potential to enhance the accuracy and allow for real-time pathological speech assessment, thereby facilitating more targeted and effective therapeutic interventions.