Abstract
Dysarthria is a speech disorder often characterized by slow speech with reduced intelligibility. Automated assessment of the severity-level and intelligibility of dysarthric speech can improve the efficiency and reliability of clinical assessment as well as benefit automatic speech recognition systems (ASR). However, in order to evaluate them, there are not sentence-level severity and intelligibility label. We only have access to speaker-per-level severity and intelligibility labels. This is a problem as dysarthric talkers might be able to produce some intelligible utterances due to frequent use and short utterances. Therefore, label based analysis might not be very accurate. To address this problem, we explore methods to estimate the severity-level and speech intelligibility in dysarthria given discrete speaker-level labeling in the training set. To accomplish this, we propose a machine learning based method using one-dimensional Convolutional Neural Networks (1-D CNN). The TORGO dataset is used to test the performance of the proposed method, with the UASpeech dataset used for Transfer learning (TL). To evaluate, an Averaged Ranking Score (ARS) and intelligibility probability distribution are used. Our findings demonstrate that the proposed method can assess speakers based on severity-level and intelligibility to provide a more granular analysis of factors underlying speech intelligibility deficits associated with dysarthria.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Duffy, J.R.: Motor speech disorders E-Book: Substrates, differential diagnosis, and management. Elsevier Health Sciences (2019)
Mitchell, C., et al.: Interventions for dysarthria due to stroke and other adult-acquired, non-progressive brain injury. Cochrane Database Syst. Rev. 1, CD002088–CD002088 (2007)
Enderby, P.: Frenchay dysarthria assessment. Br. J. Disord. Commun. 15(3), 165–173 (1980)
Dorsey, M., et al.: Speech intelligibility test for windows. Lincoln, NE: Institute for Rehabilitation Science and Engineering at Madonna Rehabilitation Hospital (2007)
Freed, D.: Motor speech disorders: diagnosis and treatment. Nelson Education (2011)
Hijikata, N., et al.: Assessment of dysarthria with Frenchay dysarthria assessment (FDA-2) in patients with Duchenne muscular dystrophy. Disabil. Rehabil., 1–8 (2020)
Kent, R.D.: Hearing and believing: some limits to the auditory-perceptual assessment of speech and voice disorders. Am. J. Speech Lang. Pathol. 5(3), 7–23 (1996)
Berisha, V., Utianski, R., Liss, J.: Towards a clinical tool for automatic intelligibility assessment. In: IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 2825–2828 (2013)
Kim, M.J., Kim, H.: Automatic assessment of dysarthric speech intelligibility based on selected phonetic quality features. In: International Conference on Computers for Handicapped Persons, pp. 447–450 (2012)
Hummel, R., Chan, W.-Y., Falk, T.H.: Spectral features for automatic blind intelligibility estimation of spastic dysarthric speech. In: Twelfth Annual Conference of the International Speech Communication Association (2011)
Ferrier, L., et al.: Dysarthric speakers’ intelligibility and speech characteristics in relation to computer speech recognition. Augment. Altern. Commun. 11(3), 165–175 (1995)
Martínez, D., et al.: Intelligibility assessment and speech recognizer word accuracy rate prediction for dysarthric speakers in a factor analysis subspace. ACM Transactions on Accessible Computing (TACCESS) 6(3), 1–21 (2015)
Gurugubelli, K., Vuppala, A.K.: Perceptually enhanced single frequency filtering for dysarthric speech detection and intelligibility assessment. In: International Conference on Acoustics, Speech and Signal Processing, pp. 6410–6414 (2019)
Bhat, C., Vachhani, B., Kopparapu, S.K.: Automatic assessment of dysarthria severity level using audio descriptors. In: International Conference on Acoustics, Speech and Signal Processing, pp. 5070–5074 (2017)
Looze, C.D., et al.: Pitch declination and reset as a function of utterance duration in conversational speech data. In: Sixteenth Annual Conference of the International Speech Communication Association (2015)
Teodorescu, H.-N.: Pitch analysis of dysarthria helps differentiating between dysarthria mechanisms. Bull. Integr. Psychiatry 84(1), 89–95 (2019)
Feenaughty, L., et al.: Speech and pause characteristics in multiple sclerosis: a preliminary study of speakers with high and low neuropsychological test performance. Clin. Linguist. Phon. 27(2), 134–151 (2013)
Allison, K.M., Yunusova, Y., Green, J.R.: Shorter sentence length maximizes intelligibility and speech motor performance in persons with dysarthria due to amyotrophic lateral sclerosis. Am. J. Speech Lang. Pathol. 28(1), 96–107 (2019)
Patel, R.: Prosodic control in severe dysarthria. J. Speech Lang. Hear. Res. 45, 858–878 (2002)
Bunton, K., et al.: Perceptuo-acoustic assessment of prosodic impairment in dysarthria. Clin. Linguist. Phon. 14(1), 13–24 (2000)
Bigi, B., et al.: A syllable-based analysis of speech temporal organization: a comparison between speaking styles in dysarthric and healthy populations. In: Sixteenth Annual Conference of the International Speech Communication Association, vol. 1, pp. 2977–2981 (2015)
Bhat, C., Strik, H.: Automatic assessment of sentence-level dysarthria intelligibility using BLSTM. J. Sel. Top. Sign. Process. 14(2), 322–330 (2020)
Joshy, A.A., Rajan, R.: Automated dysarthria severity classification using deep learning frameworks. In: European Signal Processing Conference, pp. 116–120 (2021)
Kim, J., et al.: Automatic intelligibility classification of sentence-level pathological speech. Comput. Speech Lang. 29(1), 132–144 (2015)
Kiranyaz, S., et al.: 1D convolutional neural networks and applications: a survey. In: Mechanical Systems and Signal Processing, vol. 151, p. 107398 (2021)
Rudzicz, F., Namasivayam, A.K., Wolff, T.: The TORGO database of acoustic and articulatory speech from speakers with dysarthria. Lang. Resour. Eval. 46(4), 523–541 (2012)
Kim, H., et al.: Dysarthric speech database for universal access research. In: Ninth Annual Conference of the International Speech Communication Association (2008)
Kent, R.D., et al.: Toward phonetic intelligibility testing in dysarthria. J. Speech Hear. Disord. 54(4), 482–499 (1989)
Acknowledgments
This work was supported by National Institutes of Health under NIDCD R15 DC017296-01.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Soleymanpour, M., Johnson, M.T., Berry, J. (2021). Increasing the Precision of Dysarthric Speech Intelligibility and Severity Level Estimate. In: Karpov, A., Potapova, R. (eds) Speech and Computer. SPECOM 2021. Lecture Notes in Computer Science(), vol 12997. Springer, Cham. https://doi.org/10.1007/978-3-030-87802-3_60
Download citation
DOI: https://doi.org/10.1007/978-3-030-87802-3_60
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-87801-6
Online ISBN: 978-3-030-87802-3
eBook Packages: Computer ScienceComputer Science (R0)