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

Abstract

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.

Towards robust voice pathology detection

“…Nevertheless, we consider these results exploratory due to the limitations of the databases. Reviewing the performances achieved in scenarios with MFCC as input data we conclude that MFCC alone are not reliable enough for robust voice pathology detection, which was also concluded by Ali et al in [5]. Regarding the DenseNet, we conclude that in voice pathology detection scenarios with this little training data it is better to use inputs with reduced dimensionality in contrary to raw waveform inputs, or make use of transfer learning or data augmentation.…”

“…After the feature extraction, multiple conventional classifiers have been used to detect the presence of voice pathology. Most authors relied on the following algorithms: Support Vector Machines (SVM), Gaussian Mixture Models (GMM), Random Forests (RF), and Artificial Neural Networks (ANN) [25,5,7,14], etc.…”

Edge-centric multimodal authentication system using encrypted biometric templates

“…The face images are taken from the Database of Faces [39], which was developed between 1992 and 1994 and was formerly known as the ORL Database of Faces. [40][41][42][43]. The database contains the speech signals of normal persons and dysphonic patients.…”

Recognition of pathological voices by Human Factor Cepstral Coefficients (HFCC)

“…The second module is the classifier used to classify vocal pathologies. we used a hidden Markov model with a Gaussian mixture density (HMM-GM) [9], through The Hidden Markov Model Toolkit (HTK) (HTK 3.4.1) [10] Researchers in this field have frequently used objective assessment of vocal pathology using several databases. We note here the most used databases such as the database (MEEI) [11,3], Saarbruecken Voice Database (SVD) [4,7,12] and Arabic Voice Pathology Database (AVPD) [13,7].…”