Receiver operating characteristics measure for the recognition of stuttering dysfluencies using line spectral frequencies
Stuttering is a motor-speech disorder that has features in common with other motor control disorders such as dystonia, Parkinson’s disease, and Tourette’s syndrome. Stuttering results from complex interactions between factors such as motor, language, emotions, and genetic systems. This study used Li...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English English |
| Published: |
Kulliyyah of Engineering, International Islamic University Malaysia (IIUM)
2017
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/57382/ http://irep.iium.edu.my/57382/ http://irep.iium.edu.my/57382/1/p.sed.iium_ej.2017.stuttering_dysfluencies.pdf http://irep.iium.edu.my/57382/7/57382_Receiver%20operating%20characteristics%20measure%20for%20the%20recognition%20of%20stuttering%20dysfluencies%20using%20line%20spectral%20frequencies_SCOPUS.pdf |
| Summary: | Stuttering is a motor-speech disorder that has features in common with other motor control disorders such as dystonia, Parkinson’s disease, and Tourette’s syndrome. Stuttering results from complex interactions between factors such as motor, language, emotions, and genetic systems. This study used Line Spectral Frequency (LSF) for feature extraction, while using three classifiers for the identification purpose, Multilayer Perceptron (MLP), Recurrent Neural Network (RNN) and Radial Basis Function (RBF). The UCLASS (University College London Archive of Stuttered Speech) release 1 was used as the database in this research. These recordings were from people of ages ranging from 12y11m to 19y5m, who were referred to clinics in London for assessment of their stuttering. The performance metrics used for interpreting the results are sensitivity, accuracy, precision, and misclassification rate. Only M1 and M2 had below 100% sensitivity for RBF. The sensitivity of M1 was found to be between 40% & 60%, therefore categorized as moderate, while that of M2 falls between 60% & 80%, classed as substantial. Overall, RBF outperforms the two other classifiers, MLP and RNN for all the performance metrics considered.
Gagap adalah gangguan motor pertuturan, mempunyai ciri-ciri yang sama
dengan lain-lain gangguan kawalan motor seperti dystonia, penyakit Parkinson dan
sindrom Tourette. Keputusan kegagapan daripada interaksi kompleks antara faktor-faktor
seperti motor, bahasa, emosi dan genetik. Kajian ini menggunakan Frekuensi Line
spektral (LSF) untuk pengekstrakan ciri, semasa menggunakan tiga penjodoh untuk
tujuan mengenal pasti, Multilayer Perceptron (MLP), Rangkaian Neural Berulang (RNN)
dan Radial Asas Fungsi (RBF). The UCLASS (University College London Arkib
Stuttered Ucapan) melepaskan 1 digunakan sebagai pangkalan data dalam kajian ini. Ini
rakaman adalah dari orang-orang peringkat umur 12y11m untuk 19y5m, yang dirujuk
kepada klinik di London untuk penilaian kegagapan mereka. Metrik prestasi yang
digunakan untuk mentafsir keputusan yang sensitif, ketepatan, ketepatan dan kadar
misclassification. Hanya M1 dan M2 mempunyai di bawah 100% kepekaan untuk RBF.
Kepekaan M1 didapati antara 40% & 60%, oleh itu dikategorikan sebagai sederhana,
manakala M2 jatuh antara 60% & 80%, dikelaskan sebagai besar. Secara keseluruhan,
RBF melebihi performa dua penjodoh lain, MLP dan RNN untuk semua metrik prestasi
dipertimbangkan. |
|---|