NeuroSpeech: An open-source software for Parkinson's speech analysis

Orozco-Arroyave, Juan Rafael; Vásquez-Correa, Juan Camilo; Vargas-Bonilla, Jesús Francisco; Arora, Raman; Dehak, Najim; Nidadavolu, Phani Sankar; Christensen, Heidi; Rudzicz, Frank; Yancheva, Maria; Chinaei, Hamid R.; Vann, Alyssa; Vogler, Nikolai; Bocklet, Tobias; Cerňak, Miloš; Hannink, Julius; Nöth, Elmar

doi:10.1016/j.dsp.2017.07.004

Cited by 80 publications

(53 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Phonation features are extracted from the voiced segments. The feature set includes descriptors computed for 40 ms frames of speech, including jitter, shimmer, amplitude perturbation quotient, pitch perturbation quotient, the first and second derivatives of the fundamental frequency F 0 , and the log-energy [11].…”

Section: Speech Featuresmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of User Models Based on GMM-UBM and I-Vectors for Speech, Handwriting, and Gait Assessment of Parkinson’s Disease Patients

Vásquez-Correa

Orozco-Arroyave

Nöth

2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Parkinson's disease is a neurodegenerative disorder characterized by the presence of different motor impairments. Information from speech, handwriting, and gait signals have been considered to evaluate the neurological state of the patients. On the other hand, user models based on Gaussian mixture models -universal background models (GMM-UBM) and i-vectors are considered the state-of-the-art in biometric applications like speaker verification because they are able to model specific speaker traits. This study introduces the use of GMM-UBM and i-vectors to evaluate the neurological state of Parkinson's patients using information from speech, handwriting, and gait. The results show the importance of different feature sets from each type of signal in the assessment of the neurological state of the patients.

show abstract

Section: Speech Featuresmentioning

confidence: 99%

“…Articulation: these features model aspects related to the movements of limbs involved in the speech production. The features considered the energy content in onset segments [11]. The onset detection is based on the computation of F 0 .…”

Section: Speech Featuresmentioning

confidence: 99%

Comparison of User Models Based on GMM-UBM and I-Vectors for Speech, Handwriting, and Gait Assessment of Parkinson’s Disease Patients

Vásquez-Correa

Orozco-Arroyave

Nöth

2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…The evaluation of PD patients according to the MDS-UPDRS-III scale has shown to be suitable to assess general motor impairments of PD patients; however, the deterioration of the communication skills of the PD patients is not properly evaluated because such a scale only considers speech impairments in one of its items. A modified version of the Frenchay dysarthria assessment scale (m-FDA), which can be administered based on speech recordings was recently developed [4,8]. The scale includes several aspects of speech: respiration, lips movement, palate/velum movement, larynx, tongue, monotonicity, and intelligibility.…”

Section: M-fda Scalementioning

confidence: 99%

“…Several studies in the literature have described the speech impairments developed by PD patients in terms of four different dimensions: phonation, articulation, prosody, and intelligibility [4,5]. These feature extraction strategies have shown to be suitable to support the diagnosis process and to assess the neurological state of the patients.…”

Section: Introductionmentioning

confidence: 99%

Phonological Posteriors and GRU Recurrent Units to Assess Speech Impairments of Patients with Parkinson’s Disease

Vásquez-Correa

Garcia-Ospina

Orozco-Arroyave

et al. 2018

Text, Speech, and Dialogue

Self Cite

View full text Add to dashboard Cite

Parkinson's disease is a neurodegenerative disorder characterized by a variety of motor symptoms, including several impairments in the speech production process. Recent studies show that deep learning models are highly accurate to assess the speech deficits of the patients; however most of the architectures consider static features computed from a complete utterance. Such an approach is not suitable to model the dynamics of the speech signal when the patients pronounce different sounds. Phonological features can be used to characterize the voice quality of the speech, which is highly impaired in patients suffering from Parkinson's disease. This study proposes a deep architecture based on recurrent neural networks with gated recurrent units combined with phonological posteriors to assess the speech deficits of Parkinson's patients. The aim is to model the time-dependence of consecutive phonological posteriors, which follow the sound patterns of English phonological model. The results show that the proposed approach is more accurate than a baseline based on standard acoustic features to assess the speech deficits of the patients.

show abstract

“…A SNR é calculada através de[9] usando os sinais de fala e o ruído branco da entrada de referência, uma vez que a planta apresenta norma unitária 4. Um exemplo de aplicação da API speech-to-text da Google Inc. envolvendo a avaliação de inteligibilidade no monitoramento da evolução da síndrome de Parkinson é discutida em[23] e[24].…”

unclassified

Implementação e Avaliação de Desempenho de Algoritmos para Cancelamento Adaptativo de Ruído

Pertum¹,

Kuhn²

2019

Anais De XXXVII Simpósio Brasileiro De Telecomunicações E Processamento De Sinais

View full text Add to dashboard Cite

Resumo-Este artigo trata da implementação e da avaliação de desempenho de um sistema de cancelamento adaptativo de ruído, operando com relevantes algoritmos adaptativos da literatura [isto é, o NLMS (normalized least-mean-square) e três diferentes VSS-NLMS (variable step-size NLMS)]. Essa implementação é realizada aqui em linguagem C utilizando a placa de desenvolvimento Cortex-FM4 Starter da Cypress Semiconductors ® , a qual possui recursos para aquisição e síntese de sinais de áudio. Comparações de desempenho através de metodologias objetivas para avaliação de qualidade e inteligibilidade de sinais de fala são apresentadas e discutidas, confirmando a eficácia da implementação proposta. Palavras-chave-Algoritmo NLMS, algoritmos VSS-NLMS, cancelamento adaptativo de ruído, filtragem adaptativa.Abstract-This paper deals with the implementation and performance evaluation of an adaptive noise cancelling system, operating with relevant adaptive algorithms from the literature [i.e., the normalized least-mean-square (NLMS) and three different variable step-size NLMS (VSS-NLMS)]. Such an implementation is carried out here in C language using the Cortex-FM4 Starter kit from Cypress Semiconductors ® , which has acquisition and synthesis resources for audio signals. Performance comparisons through objective methodologies for evaluating quality and intelligibility of speech signals are presented and discussed, confirming the effectiveness of the proposed implementation.

show abstract

NeuroSpeech: An open-source software for Parkinson's speech analysis

Cited by 80 publications

References 17 publications

Comparison of User Models Based on GMM-UBM and I-Vectors for Speech, Handwriting, and Gait Assessment of Parkinson’s Disease Patients

Comparison of User Models Based on GMM-UBM and I-Vectors for Speech, Handwriting, and Gait Assessment of Parkinson’s Disease Patients

Phonological Posteriors and GRU Recurrent Units to Assess Speech Impairments of Patients with Parkinson’s Disease

Implementação e Avaliação de Desempenho de Algoritmos para Cancelamento Adaptativo de Ruído

Contact Info

Product

Resources

About