Dysarthria

About: Dysarthria is a research topic. Over the lifetime, 2402 publications have been published within this topic receiving 56554 citations.

Reorganization of Speech Production at the Motor Cortex and Cerebellum following Capsular Infarction: a Follow-up Functional Magnetic Resonance Imaging Study

Abstract: Based on clinical data, Geschwind assumed left hemisphere dominance of speech production to extend to the cortical representation of articulatory and phonatory functions at the motor cortex. This author suggested, furthermore, that the clinical observation of rapid recovery from articulatory impairments after damage to the left-sided corticobulbar tracts reflects compensatory activation of an alternative pathway involving the contralateral pre-central gyrus and its efferent projections. In order to test this hypothesis, functional magnetic resonance imaging (fMRI) was performed 4 and 35 days after stroke in a 38-year-old man who had experienced sudden speech deterioration ('dysarthric speech') concomitant with weakness of the right upper limb and the right side of the face. Computerized tomography demonstrated an ischaemic infarction within the left internal capsule. The patient fully recovered from dysarthria within 9 days. Activation of the right hemisphere analogues of Broca and Wernicke areas has been assumed to contribute to recovery from aphasia. As a further aspect of the reorganization of speech function, the present case study demonstrates for the first time by means of fMRI a selective 'shift' of the cortical representation of speech motor control to the right Rolandic cortex and the left cerebellum during restitution of articulation in a case of transient dysarthria following infarction of the left internal capsule.

Automatic Speech Recognition with Deep Neural Networks for Impaired Speech

Abstract: Automatic Speech Recognition has reached almost human performance in some controlled scenarios. However, recognition of impaired speech is a difficult task for two main reasons: data is (i) scarce and (ii) heterogeneous. In this work we train different architectures on a database of dysarthric speech. A comparison between architectures shows that, even with a small database, hybrid DNN-HMM models outperform classical GMM-HMM according to word error rate measures. A DNN is able to improve the recognition word error rate a 13 % for subjects with dysarthria with respect to the best classical architecture. This improvement is higher than the one given by other deep neural networks such as CNNs, TDNNs and LSTMs. All the experiments have been done with the Kaldi toolkit for speech recognition for which we have adapted several recipes to deal with dysarthric speech and work on the TORGO database. These recipes are publicly available.

Functional outcome assessment in dysarthria.

Abstract: Strategies for the functional assessment of communication disorders experienced by persons with dysarthria can be based on the Chronic Disabilities Model described by Nagi (1991), which considers a disorders at five different levels, ranging from pathophysiology at the level of the tissue to the societal levels of dysfunction. Outcomes can be measured at all levels of the model. For example, at the pathophysiologic level, outcomes may indicate events at the tissue level during the course of the disease, whereas, at the level of the disability, outcomes reveal the adequacy of speech production using compensatory strategies in communicative contexts, and at the societal level, they may indicate the overall degree of success a speaker has in specific real-world speaking situations. This article focuses on "functional" assessment of persons with dysarthria. Thus, assessment is viewed from the perspective of how speech and the use of speech can be measured in functional situations over time or as a result of treatment.

Dysarthria in traumatic brain injury: a breath group and intonational analysis.

Abstract: Prosodic abnormality is a common feature in the dysarthrias associated with traumatic brain injury (TBI), but very few analytic studies have been reported on the nature of the prosodic disturbances. T

Model adaptation and adaptive training for the recognition of dysarthric speech

Abstract: Dysarthria is a neurological speech disorder, which exhibits multi-fold disturbances in the speech production system of an individual and can have a detrimental effect on the speech output. In addition to the data sparseness problems, dysarthric speech is characterised by inconsistencies in the acoustic space making it extremely challenging to model. This paper investigates a variety of baseline speaker independent (SI) systems and its suitability for adaptation. The study also explores the usefulness of speaker adaptive training (SAT) for implicitly annihilating inter-speaker variations in a dysarthric corpus. The paper implements a hybrid MLLR-MAP based approach to adapt the SI and SAT systems. ALL the results reported uses UASPEECH dysarthric data. Our best adapted systems gave a significant absolute gain of 11.05% (20.42% relative) over the last published best result in the literature. A statistical analysis performed across various systems and its specific implementation in modelling different dysarthric severity sub-groups, showed that, SAT-adapted systems were more applicable to handle disfluencies of more severe speech and SI systems prepared from typical speech were more apt for modelling speech with low level of severity.