Pathology Voice Detection and Classification Using Ensemble Learning
Journal Title: International Journal of Engineering and Science Invention - Year 2018, Vol 7, Issue 8
Abstract
Voice disorder is a large phenomenon which is dramatically affecting a large number of people. These pathological conditions are caused due to various reasons and some of the reasons might be talking too much, screaming constantly, smoking which affects our vocal chords. Voice disorders can be treated properly when diagnosed early. The research work in this field generally splits in two stages: first, extraction of meaningful feature sets, and second, using these features for classification of speech recordings into healthy condition and different pathological cases. The objective is to use the discriminative features in the voice signals to detect the pathologically affected voices. Here Ensemble learning technique is used to find the types of disorder in voices. The first component is the extraction of feature vectors using Mel-frequency cepstral coefficients (MFCC), Linear predictive coding (LPC), Wavelet Packet Decomposition (WPD), Cepstral Analysis (CA), Jitter, Shimmer, Pulse, Pitch, Hormonicity, Intensity, Energy and entropy methods. The second is the classification of feature vectors using ensemble learning methods. The parameters from the voice signals are used to build the model. From the experimental results, it is observed that LibD3c classifier performs well in classifying the pathological voices.
Authors and Affiliations
Mythili J, Vijaya MS
Keywords
Related Articles
- EP ID EP397558
- DOI -
- Views 54
- Downloads 0
Climatic potential: Starting point for the design of Residential housing
The investigation evaluates the fluctuations of temperature and relative humidity of the housing study of the residential suburban-rural case during 2018; therefore, the objective of the present investigation is to deter...
Modeling of Power Line Communications for Indoor System Using MIMO coding scheme
The need to study broadband Power Line Communications (PLC) for indoor use is becoming essential with the spread of broadband technologies. This paper aims at modeling the broadband PLC channel based on indoor applicatio...
Link Failure Recover by Resilience Method in Wireless Mesh Network
wireless mesh network(WMN) is a communication network. During their operating period, the wireless mesh network may suffer from frequent link failure which results in deficient performance of network. Hence, this paper p...
DFT Study on O-H Bond Dissociation Enthalpies and Ionization Potentials of Ortho, Para and Meta Substituted Phenols: A Computational Study
The O-H bond dissociation energies [BDE(O-H)] and ionization potentials of o, p, and msubstituted phenols X-PhOH [X = F, Cl, Br, OH, CN, CH3, OCH3, CF3, NH2, and NO2) have been computed using the density functional theor...
Hierarhical Forest Custer and Heuristic Poll Classification For Mining Web Access Logs
Modern web clickstream data comprises multi levels of granularity based on the visitor behavior on its websites, making it cumbersome to scrutinize. Following the overarching idea that the event visualization should prov...