Dysarthric Speakers Research Articles

Efficacy and feasibility of a digital speech therapy for post-stroke dysarthria: protocol for a randomized controlled trial.

Dysarthria is a motor speech disorder caused by various neurological diseases, particularly stroke. Individuals with post-stroke dysarthria experience impaired speech intelligibility, communication difficulties, and a reduced quality of life. However, studies on the treatment of post-stroke dysarthria are lacking. Digital speech therapy applications have the advantages of being personalized and easily accessible. However, evidence for their efficacy is not rigorous. Moreover, no studies have investigated both the acute to subacute, and chronic phases of stroke. This study aims to investigate the efficacy and feasibility of digital speech therapy applications in addressing these gaps in dysarthria treatment. This study is a multicenter, prospective, randomized, evaluator-blinded non-inferiority trial. We aim to recruit 76 participants with post-stroke dysarthria. Eligible participants will be stratified based on the onset period of stroke into acute to subacute, and chronic phases. Participants will be randomized in a 1:1 to receive either a personalized digital speech therapy application or conventional therapy with a workbook for 60 min daily, 5 days a week, for 4 weeks. The primary outcome is the improvement in speech intelligibility. This will be measured by how accurately independent listeners can transcribe passages read by the participants. Secondary outcomes, which include speech function, will be evaluated remotely by speech-language pathologists. This includes the maximum phonation time, oral diadochokinetic rate, and percentage of consonants correct. Participants' psychological well-being will also be assessed using self-report questionnaires, such as depressive symptoms (Patient Health Questionnaire-9) and quality of life (Quality of Life in the Dysarthric Speaker scale). The trial will also assess the feasibility, participant adherence, and usability of the application. Rigorous data collection and monitoring will be implemented to ensure patient safety. This trial aims to investigate the efficacy and feasibility of digital speech therapy applications for treating post-stroke dysarthria. The results could establish foundational evidence for future clinical trials with larger sample sizes. Clinicaltrials.gov, identifier: NCT05865106.

Dialect identification, intelligibility ratings, and acceptability ratings of dysarthric speech in two American English dialects

ABSTRACT The current study explored the intelligibility and acceptability ratings of dysarthric speakers with African American English (AAE) and General American English (GAE) dialects by listeners who identify as GAE or AAE speakers, as well as listener ability to identify dialect in dysarthric speech. Eighty-six listeners rated the intelligibility and acceptability of sentences extracted from a passage read by speakers with dysarthria. Samples were used from the Atlanta Motor Speech Disorders Corpus and ratings were collected via self-report. The listeners identified speaker dialect in a forced-choice format. Listeners self-reported their dialect and exposure to AAE. AAE dialect was accurately identified in 63.43% of the the opportunities; GAE dialect was accurately identified in 70.35% of the opportunities. Listeners identifying as AAE speakers rated GAE speech as more acceptable, whereas, listeners identifying as GAE speakers rated AAE speech as more acceptable. Neither group of listeners demonstrated a difference in intelligibility ratings. Exposure to AAE had no effect on intelligibility or acceptability ratings. Listeners can identify dialect (AAE and GAE) with a better than chance degree of accuracy. One’s dialect may have an effect on intelligibility and acceptability ratings. Exposure to a dialect did not affect listener ratings of intelligibility or acceptability.

Improving the intelligibility of dysarthric speech using a time domain pitch synchronous-based approach

Dysarthria is a motor speech impairment that reduces the intelligibility of speech. Observations indicate that for different types of dysarthria, the fundamental frequency, intensity, and speech rate of speech are distinct from those of unimpaired speakers. Therefore, the proposed enhancement technique modifies these parameters so that they fall in the range for unimpaired speakers. The fundamental frequency and speech rate of dysarthric speech are modified using the time domain pitch synchronous overlap and add (TD-PSOLA) algorithm. Then its intensity is modified using the fast Fourier transform (FFT) and inverse fast Fourier transform (IFFT)-based approach. This technique is applied to impaired speech samples of ten dysarthric speakers. After enhancement, the intelligibility of impaired and enhanced dysarthric speech is evaluated. The change in the intelligibility of impaired and enhanced dysarthric speech is evaluated using the rating scale and word count methods. The improvement in intelligibility is significant for speakers whose original intelligibility was poor. In contrast, the improvement in intelligibility was minimal for speakers whose intelligibility was already high. According to the rating scale method, for diverse speakers, the change in intelligibility ranges from 9% to 53%. Whereas, according to the word count method, this change in intelligibility ranges from 0% to 53%.

Sensitivity and specificity of the Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised to detect dysarthria in individuals with amyotrophic lateral sclerosis.

Given the widespread use of the Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R) to measure disease progression in ALS and recent reports demonstrating its poor sensitivity, we aimed to determine the sensitivity and specificity of the ALSFRS-R bulbar subscale and speech item to detect validated clinical ratings of dysarthria in individuals with ALS. Paired ALSFRS-R and validated Speech Intelligibility Test (SIT) data from individuals with ALS were analyzed. Trained raters completed duplicate, independent, and blinded ratings of audio recordings to obtain speech intelligibility (%) and speaking rate (words per minute, WPM). Binary dysarthria profiles were derived (dysarthria ≤96% intelligible and/or <150 WPM). Data were obtained using the Kruskal-Wallis test, receiver-operating characteristic (ROC) curve, area under the curve (AUC), sensitivity and specificity percentages, and positive/negative predictive values (PPV/NPV). A total of 250 paired SIT and ALSFRS-R data points were analyzed. Dysarthria was confirmed in 72.4% (n = 181). Dysarthric speakers demonstrated lower ALSFRS-R bulbar subscale (8.9 vs. 11.2) and speech item (2.7 vs. 3.7) scores (P < .0001). The ALSFRS-R bulbar subscale score had an AUC of 0.81 (95% confidence interval [CI] 0.75 to 0.86). A subscale score of ≤11 yielded a sensitivity of 86%, specificity of 57%, PPV of 84%, and NPV of 60% to correctly identify dysarthria status. The ALSFRS-R speech item score demonstrated an AUC of 0.81 to detect dysarthria (95% CI 0.76 to 0.85), with sensitivity of 79%, specificity of 75%, PPV of 89%, and NPV of 58% for a speech item cutpoint of ≤3. The ALSFRS-R bulbar and speech item subscale scores may be useful, inexpensive, and quick tools for monitoring dysarthria status in ALS.

Automatic speaker verification system for dysarthric speakers using prosodic features and out-of-domain data augmentation

A communication disorder is an impairment of a person’s ability to talk or communicate appropriately. Dysarthria is a common neuro-motor speech communication disorder that can be caused by neurological damage. Dysarthria may affect the articulation, phonation, and prosody of a speaker. Dysarthria patients have poor neuromotor coordination and other physical impairments, making it difficult to utilize an interactive keyboard or other user interfaces. The ASV system can make biometric applications more accessible to dysarthric speakers by eliminating the need for them to remember cumbersome and unique authentication numbers and passwords. In this paper, we presented a study on developing an automatic speaker verification (ASV) system for dysarthria patients with varying speech intelligibility to assist them in remote access control and voice-based biometric applications. In the initial part of our proposed approach, we included a duration modification-based data augmentation module in the front end of the ASV system. Since prosody deficits are one of the early indicators of dysarthria, we investigated the role of prosodic variables in combination with the traditional Mel-frequency cepstral coefficients (MFCC). The prosodic variables explored in this study include pitch, loudness, and voicing probability. Separate i-vector and x-vector models are trained and compared using individual MFCC, prosodic variables, and their combinations. The experimental results showed that the proposed approach based on combining MFCC and prosody features along with duration-modification-based data augmentation produced promising results.

Assessing speech intelligibility of pathological speech in sentences and word lists: The contribution of phoneme-level measures.

Speech intelligibility is an important indicator of the degree of speech impairment in pathological speech. Articulation, as a key feature of dysarthria, has been found to be a stronger contributor to intelligibility of dysarthric speech compared to voice quality, nasality, and prosody. In fact, therapy addressing articulation is often used by speech-language pathologists. Since phoneme-level measures are more directly related to articulation, they may contribute to better evaluating articulation imprecision in speakers with dysarthria and to monitoring the effectiveness of therapy. We collected two types of phoneme-level measures: a) Accuracy of Phonemes, the percentage of correctly transcribed phonemes, and b) Phonetic Distance, from orthographic transcriptions obtained from expert raters in two types of speech materials (i.e., meaningful sentences and word lists). We first examined the measures' interrater reliability using Generalizability Theory. Then we studied the validity of the measures by correlating them to three criterion variables. Following this, we explored their ability in distinguishing speakers in two classification tasks according to speakers' types (i.e., healthy vs dysarthric) and their severity levels of dysarthria, respectively. The results showed that both types of phoneme-level measures are highly reliable and valid in two different speech materials. They also showed acceptable results for both classification tasks in different speech materials, with word lists performing better than meaningful sentences. The differences between the two speech materials may be largely caused by differences in word structures and contextual cues in the materials. The results indicate that both types of phoneme-level measures show largely similar reliability and validity in both speech materials. These measures perform better in word lists than in meaningful sentences, suggesting an advantage for using word lists in clinical practice and research. On the other hand, meaningful sentences can be used for classifying healthy and dysarthric speakers. Our results suggest that using different speech materials gives a better overview of the speakers' intelligibility at the segmental level and the implications of their articulation impairments.

Influence of noise on perception of different types of dysarthric speech

Disturbances in the central nervous system may lead to changes in the speech muscle system, which impact muscle strength, tone, and coordination as well as movement speed, resulting in different types of dysarthria, each impacting respiration, phonation, articulation, resonance, and prosody differentially. Bilateral upper motor neuron damage leads to spastic dysarthria, cerebellar damage ataxic dysarthria, and damage to basal ganglia hypokinetic dysarthria. In estimating the severity of speech impairment, and measuring speech intelligibility, both reading and spontaneous speech samples are routinely used by speech-language pathologists. However, noise levels in acute and rehabilitation hospitals often exceed the World Health Organization guidelines (Busch-Vishniac et al., 2005) which may impact the clinical outcomes. In this study, 100 graduate student clinicians in Speech-Language Pathology measured speech intelligibility and estimated the severity of speech impairment of speech samples produced by three types of dysarthric speakers: spastic, ataxic, and hypokinetic, in two different noise conditions (quiet, and hospital noise). The differential impact of noise on different types of dysarthria is examined, and its implication on clinical outcomes is discussed.

Dysarthria detection based on a deep learning model with a clinically-interpretable layer.

Studies have shown deep neural networks (DNN) as a potential tool for classifying dysarthric speakers and controls. However, representations used to train DNNs are largely not clinically interpretable, which limits clinical value. Here, a model with a bottleneck layer is trained to jointly learn a classification label and four clinically-interpretable features. Evaluation of two dysarthria subtypes shows that the proposed method can flexibly trade-off between improved classification accuracy and discovery of clinically-interpretable deficit patterns. The analysis using Shapley additive explanation shows the model learns a representation consistent with the disturbances that define the two dysarthria subtypes considered in this work.

Comparing the performance of classic voice-driven assistive systems for dysarthric speech

Voice-driven communication assistive systems—speech enhancement (SE), voice conversion (VC), and automatic speech recognition with text-to-speech (ASR-TTS)—are recognized approaches for improving dysarthric speakers’ speech intelligibility. However, which approach performs better for moderate dysarthric patients is unclear. This study compared the benefits of three classic difference-type voice-driven assistive systems for dysarthric patients under identical test conditions. The benefits of the three systems for dysarthric patients’ speech intelligibility were compared; 14 mild-to-severedysarthric patients and five speakers with normal speech were invited to record the training sets for these systems. Five moderate dysarthric patients were selected to record two additional testing sets, which were used for evaluating the systems’ benefits. Google Automatic Speech Recognition’s (Google ASR) evaluation metrics and listening tests verified each system’s speech intelligibility and quality. The speech intelligibility results produced by Google ASR were 7.0%, 22.9%, and 93.8% for the SE, VC, and ASR-TTS systems, respectively. Regarding the listening test, the performance of speech intelligibility and quality were 38.7%, 40.5%, 95.5%, and 1.81, 2.18, 4.56 for SE, VC, and ASR-TTS systems, respectively. The ASR-TTS system performed better than SE and VC. Furthermore, t-distributed stochastic neighbor embedding (t-SNE) analysis was used to additionally compare the differences between the systems. The t-SNE analysis results indicated that ASR-TTS’ phonetic posteriorgram features provided stable performance compared with the other speech features (log-power spectrum and spectra) in the SE and VC systems. Results showed that the ASR-TTS is a potential system to improve moderate dysarthric patients’ speech intelligibility and quality in future applications.

Transcranial direct current stimulation over the primary motor cortex improves speech production in post-stroke dysarthric speakers: A randomized pilot study.

PurposeThe current study investigated the therapeutic potential of transcranial direct current stimulation (tDCS) on speech intelligibility, speech-related physiological and vocal functions among post-stroke dysarthric patients.MethodNine chronic post-stroke dysarthric patients were randomly assigned to the stimulation or sham group. The stimulation group received 2mA of anodal tDCS over the left inferior primary motor cortex for 15 minutes, while the sham group received 30s of stimulation under the same settings. All the participants received 10 daily 15 minutes of individualized speech therapy targeting their dominant phonological process or phonemes with the greatest difficulty. The outcome measures included (1) perceptual analysis of single words, passage reading and diadochokinetic rate, (2) acoustic analysis of a sustained vowel, and (3) kinematic analysis of rapid syllable repetitions and syllable production in sentence, conducted before and after the treatment.ResultsThe results revealed that both the stimulation and sham groups had improved perceptual speech intelligibility at the word level, reduced short rushes of speech during passage reading, improved rate during alternating motion rate, AMR-kha1, and improved articulatory kinematics in AMR-tha1 and syllables /tha1/ and /kha1/ production in sentence. Compared to the sham group, the stimulation group showed significant improvement in articulatory kinematics in AMR-kha1 and syllable /kha1/ production in sentence. The findings also showed that anodal stimulation led to reduced shimmer value in sustained vowel /a/ phonation, positive changes in articulatory kinematics in AMR-tha1 and syllables /pha1/ and /kha1/ production in sentence at the post treatment measure. In addition to positive effects on articulatory control, reduced perturbation of voice amplitude documented in the stimulation group post treatment suggests possible tDCS effects on the vocal function.ConclusionsThe current study documented the beneficial effects of anodal tDCS over the primary motor cortex on speech production and suggested that combined tDCS and speech therapy may promote recovery from post-stroke dysarthria.

An Investigation to Identify Optimal Setup for Automated Assessment of Dysarthric Intelligibility using Deep Learning Technologies

Recent advances in deep learning have provided an opportunity to improve and automate dysarthria intelligibility assessment, offering a cost-effective, accessible, and less subjective way to assess dysarthric speakers. However, reviewing previous literature in the area determines that the generalization of results on new dysarthric patients was not measured properly or incomplete among the previous studies that yielded very high accuracies due to the gaps in the adopted evaluation methodologies. This is of particular importance as any practical and clinical application of intelligibility assessment approaches must reliably generalize on new patients; otherwise, the clinicians cannot accept the assessment results provided by the system deploying the approach. In this paper, after these gaps are explained, we report on our extensive investigation to propose a deep learning–based dysarthric intelligibility assessment optimal setup. Then, we explain different evaluation strategies that were applied to thoroughly verify how the optimal setup performs with new speakers and across different classes of speech intelligibility. Finally, a comparative study was conducted, benchmarking the performance of our proposed optimal setup against the state of the art by adopting similar strategies previous studies employed. Results indicate an average of 78.2% classification accuracy for unforeseen low intelligibility speakers, 40.6% for moderate intelligibility speakers, and 40.4% for high intelligibility speakers. Furthermore, we noticed a high variance of classification accuracies among individual speakers. Finally, our proposed optimal setup delivered an average of 97.19% classification accuracy when adopting a similar evaluation strategy used by the previous studies.

Acoustic Change Over Time in Spastic and/or Flaccid Dysarthria in Motor Neuron Diseases

This study aims to investigate acoustic change over time as biomarkers to differentiate among spastic-flaccid dysarthria associated with amyotrophic lateral sclerosis (ALS), spastic dysarthria associated with primary lateral sclerosis (PLS), flaccid dysarthria associated with spinal and bulbar muscular atrophy (SBMA), and to explore how these acoustic parameters are affected by dysarthria severity. Thirty-three ALS patients with mixed flaccid-spastic dysarthria, 17 PLS patients with pure spastic dysarthria, 18 SBMA patients with pure flaccid dysarthria, and 70 controls, all French speakers, were included in the study. Speakers produced vowel-glide sequences targeting different vocal tract shape changes. The mean and coefficient of variation of the total squared change of mel frequency cepstral coefficients were used to capture the degree and variability of acoustic changes linked to vocal tract modifications over time. Differences in duration of acoustic events were also measured. All pathological groups showed significantly less acoustic change compared to controls, reflecting less acoustic contrast in sequences. Spastic and mixed spastic-flaccid dysarthric speakers showed smaller acoustic changes and slower sequence production compared to flaccid dysarthria. For dysarthria subtypes associated with a spastic component, reduced degree of acoustic change was also associated with dysarthria severity. The acoustic parameters partially differentiated among the dysarthria subtypes in relation to motor neuron diseases. While similar acoustic patterns were found in spastic-flaccid and spastic dysarthria, crucial differences were found between these two subtypes relating to variability. The acoustic patterns were much more variable in ALS. This method forms a promising clinical tool as a diagnostic marker of articulatory impairment, even at mild stage of dysarthria progression in all subtypes.

A serious game for speech training in dysarthric speakers with Parkinson's disease: Exploring therapeutic efficacy and patient satisfaction.

BackgroundThe increasing need for speech therapy due to our ageing population raises the demand on therapeutical resources. To meet this demand, innovative delivery of speech training is required. eHealth applications may provide a solution, as intensified and prolonged training is only possible and affordable in patients’ home environment.AimsThis study explores the effects on speech intelligibility of game‐based speech training that provides automatic feedback on loudness, pitch and pronunciation. Additionally, we investigate how satisfied patients are with the game‐based speech training and how they experience the automatic feedback. Furthermore, patients’ preferences for game‐based speech training compared with face‐to‐face training are explored.Methods and ProceduresEight adult dysarthric speakers with Parkinson's disease (PD) completed a 4‐week game‐based speech training in their home environment. For each speaker, 24 speech utterances were audio recorded 4 weeks before (T1), immediately before (T2) and immediately after (T3) the training. All speech samples were rated on speech intelligibility by 10 untrained listeners, by comparing them with the corresponding utterances realized by a healthy speaker. Changes over time were analysed using a linear mixed‐effects analysis. Patient satisfaction with the game and the automatic feedback was assessed using a questionnaire. The preferences of patients were collected using a paired comparisons procedure in which the patients were asked whether they would prefer game‐based or face‐to‐face speech training in four hypothetical scenarios with different hypothesized levels of speech improvement.Outcomes and ResultsWhile there was no significant difference in speech intelligibility ratings between T1 and T2, we did find one between T2 and T3. At T3, speech intelligibility was rated higher than at T2, indicating positive effects of the game‐based speech training. Patients generally seemed satisfied with the game as average ratings were above 7 on a 10‐point rating scale. Generally, patients agreed with the automatic feedback and could use it to positively change the way they spoke. Patients prefer the training that provides the highest hypothetical improvement, and thus do not prefer face‐to‐face above game‐based therapy.Conclusions and ImplicationsThe results of this study suggest that dysarthric speakers due to PD see game‐based speech therapy as a valid alternative for face‐to‐face therapy and that it leads to an average improvement in speech intelligibility. For an optimal effect and user satisfaction it should preferably not be used in isolation but in combination with face‐to‐face training. In this manner, the strengths of both therapeutic deliveries can be harnessed.WHAT THIS PAPER ADDSWhat is already known on this subjectDysarthric speech in patients with PD is known to benefit from intensified and long‐term speech therapy. The increasing need for speech therapy due to our ageing population raises the demand on therapeutical resources making highly frequent and long‐term therapy difficult. eHealth provides the opportunity to intensify and prolong speech training in patients’ home environment. A drill‐and‐practice method was employed and investigated in a web‐based speech application, indicating positive effects on speech intelligibility. However, participants indicated a lack of variation in exercises making the training less enjoyable. Other research showed that serious games can increase enjoyment during training.What this paper adds to existing knowledgeThe results of this study show that it is possible to develop a serious game that can be successfully used for speech training by patients with dysarthria due to PD. Comparing speech intelligibility ratings before and after training, we observed significant improvements in speech intelligibility ratings. Patients generally agree with the automatic feedback and can use it to positively change the way they speak. Average ratings were above 7 on a 10‐point scale, indicating that patients are satisfied with the game. Patients prefer the type of training (game based or face to face) that provides the highest hypothetical improvement.What are the potential or actual clinical implications of this work?The results of this study suggest that game‐based speech training can improve speech intelligibility in patients. This indicates that it can be considered a suitable approach in the treatment of patients with dysarthria due to PD. Patients do not prefer game‐based training above face‐to‐face training in all scenarios. For that reason, for every patient one should carefully consider how to optimally combine game‐based and face‐to‐face training.

Orofacial Muscle Strength across the Dysarthrias.

This study compared orofacial muscle strength between normal and dysarthric speakers and across types of dysarthria, and examined correlations between strength and dysarthria severity. Participants included 79 speakers with flaccid, spastic, mixed spastic–flaccid, ataxic, or hypokinetic dysarthria and 33 healthy controls. Maximum pressure generation (Pmax) by the tongue, lips, and cheeks represented strength. Pmax was lower for speakers with mixed spastic–flaccid dysarthria for all tongue and lip measures, as well as for speakers with flaccid or spastic dysarthria for anterior tongue elevation and lip compression. Anterior tongue elevation and cheek compression tended to be lower than normal for the hypokinetic group. Pmax did not differ significantly between controls and speakers with ataxic dysarthria on any measure. Correlations were generally weak between dysarthria severity and orofacial weakness but were stronger in the dysarthria groups with more prominent orofacial weakness. The results generally support predictions that orofacial weakness accompanies flaccid and/or spastic dysarthria but not ataxic dysarthria. The findings support including type of dysarthria as a variable of interest when examining orofacial weakness in motor speech disorders.

Robust HI and dysarthric speaker recognition – perceptual features and models

Robust HI and dysarthric speaker recognition – perceptual features and models

Speaker Adaptation Using Spectro-Temporal Deep Features for Dysarthric and Elderly Speech Recognition

Despite the rapid progress of automatic speech recognition (ASR) technologies targeting normal speech in recent decades, accurate recognition of dysarthric and elderly speech remains highly challenging tasks to date. Sources of heterogeneity commonly found in normal speech including accent or gender, when further compounded with the variability over age and speech pathology severity level, create large diversity among speakers. To this end, speaker adaptation techniques play a key role in personalization of ASR systems for such users. Motivated by the spectro-temporal level differences between dysarthric, elderly and normal speech that systematically manifest in articulatory imprecision, decreased volume and clarity, slower speaking rates and increased dysfluencies, novel spectro-temporal subspace basis deep embedding features derived using SVD speech spectrum decomposition are proposed in this paper to facilitate auxiliary feature based speaker adaptation of state-of-the-art hybrid DNN/TDNN and end-to-end Conformer speech recognition systems. Experiments were conducted on four tasks: the English UASpeech and TORGO dysarthric speech corpora; the English DementiaBank Pitt and Cantonese JCCOCC MoCA elderly speech datasets. The proposed spectro-temporal deep feature adapted systems outperformed baseline i-Vector and x-Vector adaptation by up to 2.63&#x0025; absolute (8.63&#x0025; relative) reduction in word error rate (WER). Consistent performance improvements were retained after model based speaker adaptation using learning hidden unit contributions (LHUC) was further applied. The best speaker adapted system using the proposed spectral basis embedding features produced the lowest published WER of 25.05&#x0025; on the UASpeech test set of 16 dysarthric speakers.

Translation and validation of communicative quality of life in the dysarthric speaker questionnaire in telugu

Purpose: Several tools are available to assess the quality of life for speech, language, voice, and swallowing disorders in Indian languages. However, limited tools are present to assess the quality of life of dysarthria. So, the present study aims to translate and validate the Quality of Life in the Dysarthric Speaker (QoL-DyS) questionnaire into the Telugu language.Methods: A total of 30 different types of dysarthria and 30 age-matched controls participated in the present study. The QoL-DyS tool was trans-adapted in the Telugu language using a standardized procedure and administered to the participants. The internal consistency, test-retest reliability, construct validity, scale statistics, and item statistics were computed.Results: The present study revealed that the developed tool has good (i.e., 0.81) overall internal consistency and excellent (i.e., 0.92) test-retest reliability. The significant difference (i.e., &lt;i&gt;p&lt;/i&gt;&lt;0.05) was found between the experimental and control group.Conclusions: From the present study, we can conclude that the Telugu QOL-DyS is a reliable tool. It is applicable in clinical practice for the self-assessment of the quality of life in dysarthric speakers. Hence, the Telugu QOL-DyS is recommended in clinical practice, and there is a need to develop QoL-DyS in other Indian languages.

Significant sensors and parameters in assessment of dysarthric speech

PurposeDysarthria is a neuromotor speech disorder caused by neuromuscular disturbances that affect one or more articulators resulting in unintelligible speech. Though inter-phoneme articulatory variations are well captured by formant frequency-based acoustic features, these variations are expected to be much higher for dysarthric speakers than normal. These substantial variations can be well captured by placing sensors in appropriate articulatory position. This study focuses to determine a set of articulatory sensors and parameters in order to assess articulatory dysfunctions in dysarthric speech.Design/methodology/approachThe current work aims to determine significant sensors and parameters associated using motion path and correlation analyzes on the TORGO database of dysarthric speech. Among eight informative sensor channels and six parameters per channel in positional data, the sensors such as tongue middle, back and tip, lower and upper lips and parameters (y, z, φ) are found to contribute significantly toward capturing the articulatory information. Acoustic and positional data analyzes are performed to validate these identified significant sensors. Furthermore, a convolutional neural network-based classifier is developed for both phone-and word-level classification of dysarthric speech using acoustic and positional data.FindingsThe average phone error rate is observed to be lower, up to 15.54% for positional data when compared with acoustic-only data. Further, word-level classification using a combination of both acoustic and positional information is performed to study that the positional data acquired using significant sensors will boost the performance of classification even for severe dysarthric speakers.Originality/valueThe proposed work shows that the significant sensors and parameters can be used to assess dysfunctions in dysarthric speech effectively. The articulatory sensor data helps in better assessment than the acoustic data even for severe dysarthric speakers.

뇌성마비 마비말장애 성인의 파찰음 실현 양상 분석*

This study aims to analyze the acoustic characteristics of Korean affricates produced by dysarthric speakers with cerebral palsy. Korean fricatives and affricates are the consonants that are prone to errors in dysarthric speech, but previous studies have focused only on fricatives. For this study, three affricates /tɕ, tɕ^h, t͈ɕ/ appearing at word initial and intervocalic positions produced by six mild-moderate male speakers of spastic dysarthria are selected from a QOLT database constructed in 2014. The parameters representing the acoustic characteristics of Korean affricates were extracted by using Praat: frication duration, closure duration, center of gravity, variance, skewness, kurtosis, and central moment. The results are as follows: 1) frication duration of the intervocalic affricates produced by dysarthric speakers was significantly longer than that of the non-disordered speakers; 2) the closure duration of dysarthric speakers was significantly longer; 3) in the case of the center of gravity, there was no significant difference between the two groups; 4) the skewness of the dysarthric speakers was significantly larger; and 5) the central moment of dysarthric speakers was significantly larger. This study investigated the characteristics of the affricates produced by dysarthric speakers and differences with non-disordered speakers.

음질, 운율, 발음 특징을 이용한 마비말장애 중증도 자동 분류*

This study focuses on the issue of automatic severity classification of dysarthric speakers based on speech intelligibility. Speech intelligibility is a complex measure that is affected by the features of multiple speech dimensions. However, most previous studies are restricted to using features from a single speech dimension. To effectively capture the characteristics of the speech disorder, we extracted features of multiple speech dimensions: voice quality, prosody, and pronunciation. Voice quality consists of jitter, shimmer, Harmonic to Noise Ratio (HNR), number of voice breaks, and degree of voice breaks. Prosody includes speech rate (total duration, speech duration, speaking rate, articulation rate), pitch (F0 mean/std/ min/max/med/25quartile/75 quartile), and rhythm (%V, deltas, Varcos, rPVIs, nPVIs). Pronunciation contains Percentage of Correct Phonemes (Percentage of Correct Consonants/Vowels/Total phonemes) and degree of vowel distortion (Vowel Space Area, Formant Centralized Ratio, Vowel Articulatory Index, F2-Ratio). Experiments were conducted using various feature combinations. The experimental results indicate that using features from all three speech dimensions gives the best result, with a 80.15 F1-score, compared to using features from just one or two speech dimensions. The result implies voice quality, prosody, and pronunciation features should all be considered in automatic severity classification of dysarthria.