Parallel Representation Learning for the Classification of Pathological Speech: Studies on Parkinson’s Disease and Cleft Lip and Palate

Vásquez-Correa, Juan Camilo; Arias-Vergara, Tomás; Schuster, Maria; Orozco-Arroyave, Juan Rafael; Nöth, Elmar

doi:10.1016/j.specom.2020.07.005

Cited by 30 publications

(25 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sustained vowels phonation is perhaps the most popular, also because it represents a very common task in different applications [ 24 ]. Other works focus on continuous speech recordings including sentences, read texts, and spontaneous speech, where clinically informative phenomena like prosody can be analyzed [ 19 , 37 , 48 ]. Few papers focused on the production of isolated words.…”

Section: Related Workmentioning

confidence: 99%

An algorithm for Parkinson’s disease speech classification based on isolated words analysis

Amato

Borzì

Olmo

et al. 2021

Health Inf Sci Syst

View full text Add to dashboard Cite

Introduction Automatic assessment of speech impairment is a cutting edge topic in Parkinson’s disease (PD). Language disorders are known to occur several years earlier than typical motor symptoms, thus speech analysis may contribute to the early diagnosis of the disease. Moreover, the remote monitoring of dysphonia could allow achieving an effective follow-up of PD clinical condition, possibly performed in the home environment. Methods In this work, we performed a multi-level analysis, progressively combining features extracted from the entire signal, the voiced segments, and the on-set/off-set regions, leading to a total number of 126 features. Furthermore, we compared the performance of early and late feature fusion schemes, aiming to identify the best model configuration and taking advantage of having 25 isolated words pronounced by each subject. We employed data from the PC-GITA database (50 healthy controls and 50 PD patients) for validation and testing. Results We implemented an optimized k-Nearest Neighbours model for the binary classification of PD patients versus healthy controls. We achieved an accuracy of 99.4% in 10-fold cross-validation and 94.3% in testing on the PC-GITA database (average value of male and female subjects). Conclusion The promising performance yielded by our model confirms the feasibility of automatic assessment of PD using voice recordings. Moreover, a post-hoc analysis of the most relevant features discloses the option of voice processing using a simple smartphone application.

show abstract

Section: Related Workmentioning

confidence: 99%

An algorithm for Parkinson’s disease speech classification based on isolated words analysis

Amato

Borzì

Olmo

et al. 2021

Health Inf Sci Syst

View full text Add to dashboard Cite

show abstract

“…We can differentiate two axes. On one side, the estimation of neurodegenerative pathologies that impact directly and typically the voice production of the subjects, such as pre-symptomatic Hungtington's disease [87% of accuracy reported in Rusz et al ( 3 )], Alzheimer's disease [80% of accuracy in Weiner et al ( 4 )], dysphonia [89% of accuracy in Tulics et al ( 5 )], or Parkinson's disease [84% of accuracy in Vasquez-Correa et al ( 6 )].…”

Section: Introductionmentioning

confidence: 99%

How to Design a Relevant Corpus for Sleepiness Detection Through Voice?

Martin

Rouas

Micoulaud‐Franchi

et al. 2021

Front. Digit. Health

View full text Add to dashboard Cite

This article presents research on the detection of pathologies affecting speech through automatic analysis. Voice processing has indeed been used for evaluating several diseases such as Parkinson, Alzheimer, or depression. If some studies present results that seem sufficient for clinical applications, this is not the case for the detection of sleepiness. Even two international challenges and the recent advent of deep learning techniques have still not managed to change this situation. This article explores the hypothesis that the observed average performances of automatic processing find their cause in the design of the corpora. To this aim, we first discuss and refine the concept of sleepiness related to the ground-truth labels. Second, we present an in-depth study of four corpora, bringing to light the methodological choices that have been made and the underlying biases they may have induced. Finally, in light of this information, we propose guidelines for the design of new corpora.

show abstract

“…automatically learn high-level discriminative dysarthric representations [9][10][11][12][13][14]. Given the potential of deep learning-based approaches to characterize abstract but important acoustic cues beyond the realm of knowledge-based handcrafted features, in this paper we focus on deep learning-based approaches.…”

Section: Introductionmentioning

confidence: 99%

“…Mainstream deep learning-based dysarthric speech detection approaches rely on processing the magnitude spectrum (or features derived from the magnitude spectrum) of time-frequency representations such as the short-time Fourier transform (STFT) or continuous wavelet transform [10][11][12][13][14]. In [10], articulation impairments of patients suffering from dysarthria are modeled through a convolutional neural network (CNN) operating on the magnitude spectrum of the continuous wavelet transform.…”

Section: Introductionmentioning

confidence: 99%

“…In [11,12], a CNN is trained on the STFT magnitude spectrum or Mel frequency cepstral coefficients of neurotypical and dysarthric input signals. The STFT magnitude spectrum is also used in [13,14] to train unsupervised and supervised auto-encoders for dysarthric speech detection. Although considerable insight about the structure of a signal can indeed be obtained from the magnitude spectrum, there are inherent structures also in the phase spectrum, which however has been largely ignored in automatic dysarthric speech detection techniques.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Experimental investigation on STFT phase representations for deep learning-based dysarthric speech detection

Janbakhshi¹,

Kodrasi²

2021

Preprint

View full text Add to dashboard Cite

Mainstream deep learning-based dysarthric speech detection approaches typically rely on processing the magnitude spectrum of the short-time Fourier transform of input signals, while ignoring the phase spectrum. Although considerable insight about the structure of a signal can be obtained from the magnitude spectrum, the phase spectrum also contains inherent structures which are not immediately apparent due to phase discontinuity. To reveal meaningful phase structures, alternative phase representations such as the modified group delay (MGD) spectrum and the instantaneous frequency (IF) spectrum have been investigated in several applications. The objective of this paper is to investigate the applicability of the unprocessed phase, MGD, and IF spectra for dysarthric speech detection. Experimental results show that dysarthric cues are present in all considered phase representations. Further, it is shown that using phase representations as complementary features to the magnitude spectrum is very beneficial for deep learning-based dysarthric speech detection, with the combination of magnitude and IF spectra yielding a very high performance. The presented results should raise awareness in the research community about the potential of the phase spectrum for dysarthric speech detection and motivate further research into novel architectures that optimally exploit magnitude and phase information.

show abstract

Parallel Representation Learning for the Classification of Pathological Speech: Studies on Parkinson’s Disease and Cleft Lip and Palate

Cited by 30 publications

References 37 publications

An algorithm for Parkinson’s disease speech classification based on isolated words analysis

An algorithm for Parkinson’s disease speech classification based on isolated words analysis

How to Design a Relevant Corpus for Sleepiness Detection Through Voice?

Experimental investigation on STFT phase representations for deep learning-based dysarthric speech detection

Contact Info

Product

Resources

About