@InProceedings{Mousavi2023_1193,
author = {Neda Mousavi and Sven Grawunder},
booktitle = {Studientexte zur Sprachkommunikation: Elektronische Sprachsignalverarbeitung 2023},
title = {Persian Speaker Classification Using Rhythmic Features},
year = {2023},
editor = {Christoph Draxler},
month = mar,
pages = {194--201},
publisher = {TUDpress, Dresden},
abstract = {We applied three common supervised classification models, including Random Forest (RF), Naive Bayes (NB), and Support Vector Machine (SVM), to classify speakers based on rhythm features. The dataset consisted of a set of read speech by 8 Persian speakers. Following previous studies, rhythm parameters in domains other than time, namely intensity and frequency, had been considered in the selection of rhythmic features and subsequently used for classifying speakers. Whereas PVIs and rate features contribute most to accuracy and Gini in particular for RF, there seems to be no aggravating difference between model schemes (OvR and OvO) with respect to performance.},
isbn = {978-3-95908-303-4},
issn = {0940-6832},
keywords = {Poster},
url = {https://www.essv.de/pdf/pdf/2023_194_201.pdf},
}