Quantification of Linear and Non-linear Acoustic Analysis Applied to Voice Pathology Detection

Panek, Daria; Skalski, Andrzej; Gajda, Janusz

doi:10.1007/978-3-319-06596-0_33

Cited by 9 publications

(8 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Female results of vowel a also show higher results up to 32% than publication [2]. Male results are more or less similar to those presented in [2]. Furthermore, it has been shown that fuzzy membership grade of a given voice in a patient cluster can be used as a tool for the voice pathology detection.…”

Section: Discussionsupporting

confidence: 63%

“…Another work shows 94,93% accuracy of detection pathological files and 95% of healthy ones [27]. Female results of vowel a also show higher results up to 32% than publication [2]. Male results are more or less similar to those presented in [2].…”

Section: Discussionmentioning

confidence: 65%

“…Speech signal analysis was performed with 28 parameters: fundamental frequency, jitter and shimmer coefficients, mean value of the signal, energy, 0-order, 1st-order, 2nd-order and 3rd-order moments, kurtosis, 1st, 2nd and 3rd formant's amplitude, 1st, 2nd and 3rd formant's frequency, maximum and minimum value of the signal and 10 mel-frequency cepstrum coefficients (MFCC). The mathematical basic of presented feature vector is described in previous paper [2]. Because, the feature vector is composed from a lot of parameters, it was necessary to extract the most significant information and reduce the number of parameters.…”

Section: Methodsmentioning

confidence: 99%

“…Because, the feature vector is composed from a lot of parameters, it was necessary to extract the most significant information and reduce the number of parameters. To do this, linear Principal Component Analysis was provided [2]. To compare linear reduction method, 2 non-linear transformation methods were used: non-linear PCA (NLPCA) based on neural networks and kernel PCA (kPCA) with the usage of Gauss kernel function.…”

Section: Methodsmentioning

confidence: 99%

See 3 more Smart Citations

Voice pathology detection by fuzzy logic

Panek

Skalski

Gajda

2015

2015 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) Proceedings

Self Cite

View full text Add to dashboard Cite

In this paper an efficient feature extraction methods and fuzzy logic based disorder assessment technique were used to investigate voice signals of patients suffering from functional dysphonia, hyperfunctional dysphonia, vocal cord paralysis and laryngitis. In this work, a vector made up from 28 acoustic parameters was an input for Principal Component Analysis, kernel Principal Component Analysis and Auto-associative Neural Network. Using S-shaped membership function of fuzzy logic, signals were clustered into 2 classes -healthy and pathology one. The amount of fuzzy membership of normal and pathological voice signals in their corresponding clusters was a measure to quantify the membership of the features of a particular class. In the end, S-shaped fuzzy logic method was used as a way of voice pathology detection. A classification accuracy up to 100 percent was achieved using initial 28 feature vector.This full text paper was peer-reviewed at the direction of IEEE Instrumentation and Measurement Society prior to the acceptance and publication.978-1-4799-6144-6/15/$31.00 ©2015 IEEE

show abstract

Section: Discussionsupporting

confidence: 63%

Section: Discussionmentioning

confidence: 65%

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

See 2 more Smart Citations

Voice pathology detection by fuzzy logic

Panek

Skalski

Gajda

2015

2015 IEEE International Instrumentation and Measurement Technology Conference (I2MTC) Proceedings

Self Cite

View full text Add to dashboard Cite

show abstract

“…Therefore, many automatic voice detection systems have been developed by researchers for the objective evaluation of vocal fold disorders [30][31][32][33]. Automatic disorder detection systems can be used by general practitioners for early screening, and in the case of a voice problem, they can refer the patient to a specialist.…”

Section: Introductionmentioning

confidence: 99%

Intra- and Inter-database Study for Arabic, English, and German Databases: Do Conventional Speech Features Detect Voice Pathology?

et al. 2017

View full text Add to dashboard Cite

A large population around the world has voice complications. Various approaches for subjective and objective evaluations have been suggested in the literature. The subjective approach strongly depends on the experience and area of expertise of a clinician, and human error cannot be neglected. On the other hand, the objective or automatic approach is noninvasive. Automatic developed systems can provide complementary information that may be helpful for a clinician in the early screening of a voice disorder. At the same time, automatic systems can be deployed in remote areas where a general practitioner can use them and may refer the patient to a specialist to avoid complications that may be life threatening. Many automatic systems for disorder detection have been developed by applying different types of conventional speech features such as the linear prediction coefficients, linear prediction cepstral coefficients, and Mel-frequency cepstral coefficients (MFCCs). This study aims to ascertain whether conventional speech features detect voice pathology reliably, and whether they can be correlated with voice quality. To investigate this, an automatic detection system based on MFCC was developed, and three different voice disorder databases were used in this study. The experimental results suggest that the accuracy of the MFCC-based system varies from database to database. The detection rate for the intra-database ranges from 72% to 95%, and that for the inter-database is from 47% to 82%. The results conclude that conventional speech features are not correlated with voice, and hence are not reliable in pathology detection.

show abstract