Evaluation of an Automatic Speech Recognition Platform for Dysarthric Speech

Calvo, Irene; Tropea, Peppino; Viganò, Mauro; Scialla, Maria; Cavalcante, Agnieszka Betkowska; Grajzer, Monika; Gilardone, Marco; Corbo, Massimo

doi:10.1159/000511042

Cited by 23 publications

(7 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…With dysarthric ASR, SD and SA approaches are more favorable due to the increased speech variability between different dysarthric individuals and the existing data scarcity problem significantly impacting the performance of SI dysarthric ASR. Among SD and SA dysarthric ASR, SA seems to be more on-demand lately as the SI system can be initially trained using normal speech and then adopt to individual speakers with dysarthria [5].…”

Section: Related Workmentioning

confidence: 99%

“…These differences make acoustic modeling components in standard ASR systems ineffective in mapping dysarthric speech signals to phonemes correctly; they need to deal with challenges caused by unusual and imprecise phonation, tempo and speed inconsistencies, sonorants random shifting of formant frequencies, etc. Thus, normal speech recognition systems have shown poor performance in recognizing dysarthric speech [5]. A review conducted in [6] suggests that while state-of-the-art normal ASR systems perform well on mild dysarthria, the performance degrades significantly as the condition worsens.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Speech Vision: An End-to-End Deep Learning-Based Dysarthric Automatic Speech Recognition System

Shahamiri

2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

Dysarthria is a disorder that affects an individual's speech intelligibility due to the paralysis of muscles and organs involved in the articulation process. As the condition is often associated with physically debilitating disabilities, not only do such individuals face communication problems, but also interactions with digital devices can become a burden. For these individuals, automatic speech recognition (ASR) technologies can make a significant difference in their lives as computing and portable digital devices can become an interaction medium, enabling them to communicate with others and computers. However, ASR technologies have performed poorly in recognizing dysarthric speech, especially for severe dysarthria, due to multiple challenges facing dysarthric ASR systems. We identified these challenges are due to the alternation and inaccuracy of dysarthric phonemes, the scarcity of dysarthric speech data, and the phoneme labeling imprecision. This paper reports on our second dysarthric-specific ASR system, called Speech Vision (SV) that tackles these challenges by adopting a novel approach towards dysarthric ASR in which speech features are extracted visually, then SV learns to see the shape of the words pronounced by dysarthric individuals. This visual acoustic modeling feature of SV eliminates phonemerelated challenges. To address the data scarcity problem, SV adopts visual data augmentation techniques, generates synthetic dysarthric acoustic visuals, and leverages transfer learning. Benchmarking with other state-of-the-art dysarthric ASR considered in this study, SV outperformed them by improving recognition accuracies for 67% of UA-Speech speakers, where the biggest improvements were achieved for severe dysarthria.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Speech Vision: An End-to-End Deep Learning-Based Dysarthric Automatic Speech Recognition System

Shahamiri

2021

IEEE Trans. Neural Syst. Rehabil. Eng.

View full text Add to dashboard Cite

show abstract

“…ej., Flácida, espástica, mixta, hipercinética) y niveles de gravedad, todos los cuales son desafíos para que los algoritmos informáticos administren de manera efectiva. Se han documentado tasas altas de reconocimiento del habla (80% y más) principalmente para aquellos con disartria leve (Calvo et al, 2020). Aunque el uso del reconocimiento de voz por parte de personas sin impedimentos del habla ha aumentado como resultado de aplicaciones basadas en el consumidor como Alexa o Siri.…”

Section: Técnicas De Reconocimiento Del Hablaunclassified

Redesign of user interface for e-government application using usability testing method

Adinda

Suzianti

2018

Proceedings of the 4th International Conference on Communication and Information Processing

View full text Add to dashboard Cite

show abstract

“…Since dysarthria is a frequent symptom of speech disorder, it has to be identified and diagnosed in the early stage itself. Assessment and classification of dysarthria gained significant importance because of the need to understand the variety of impairment results in speech disorder [4] and to develop speech and language therapy which can be used to encourage the patients to improve their communication skills. Many works had been carried out to identify the dysarthria speech and the level of severity in the dysarthria by extracting features like MFCC [5], LPC [6], Log RASTA PLP [7], Centroid Formants [8], glottal features [9], Perceptually Enhanced Single Frequency Cepstral Coefficients [10], and Spectro -temporal sparsity features extracted through STFT with Mel warping and Single Frequency Filtering [11].…”

Section: Introductionmentioning

confidence: 99%

Comparative analysis of deep learning models for dysarthric speech detection

Shanmugapriya,

Mohan

2023

Soft Comput

View full text Add to dashboard Cite

Dysarthria is a speech communication disorder that is associated with neurological impairments. In order to detect this disorder from speech, we present an experimental comparison of deep models developed based on frequency domain features. A comparative analysis of deep models is performed in the detection of dysarthria using scalogram of Dysarthric Speech. Also, it can assist physicians, specialists, and doctors based on the results of its detection. Since Dysarthric speech signals have segments of breathy and semi-whispery, experiments are performed only on the frequency domain representation of speech signals. Time domain speech signal is transformed into a 2-D scalogram image through wavelet transformation. Then, the scalogram images are applied to pre-trained convolutional neural networks. The layers of pre-trained networks are tuned for our scalogram images through transfer learning. The proposed method of applying the scalogram images as input to pre-trained CNNs is evaluated on the TORGO database and the classification performance of these networks is compared. In this work, AlexNet, GoogLeNet, and ResNet 50 are considered deep models of pre-trained convolutional neural networks. The proposed method of using pre-trained and transfer learned CNN with scalogram image feature achieved better accuracy when compared to other machine learning models in the dysarthria detection system.

show abstract

Evaluation of an Automatic Speech Recognition Platform for Dysarthric Speech

Cited by 23 publications

References 28 publications

Speech Vision: An End-to-End Deep Learning-Based Dysarthric Automatic Speech Recognition System

Speech Vision: An End-to-End Deep Learning-Based Dysarthric Automatic Speech Recognition System

Redesign of user interface for e-government application using usability testing method

Comparative analysis of deep learning models for dysarthric speech detection

Contact Info

Product

Resources

About