logo
ResearchBunny Logo
Multi-class identification of tonal contrasts in Chokri using supervised machine learning algorithms

Linguistics and Languages

Multi-class identification of tonal contrasts in Chokri using supervised machine learning algorithms

A. Gope, A. Pal, et al.

This groundbreaking study by Amalesh Gope, Anusuya Pal, Sekholu Tetseo, Tulika Gogoi, Joanna J, and Dinkur Borah explores the use of machine learning algorithms to identify tonal contrasts in the endangered Chokri language. With astonishing accuracy rates reaching 95-97% for male speakers, the research uncovers the potential of these algorithms in analyzing tonal languages, paving the way for broader applications.

00:00
00:00
Playback language: English
Abstract
This study investigates the efficacy of various machine learning algorithms (MLAs) in identifying tonal contrasts within the endangered Chokri language. Seven supervised MLAs (Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, K-Nearest Neighbors, Naive Bayes) and an Artificial Neural Network (ANN) were employed to analyze five-way tonal contrasts. Acoustic correlates, focusing on fundamental frequency (f0) height and direction, were examined. Random Forest demonstrated superior accuracy, exceeding the performance of ANN. The study also revealed that combining f0 height and direction improved recognition for female speakers, while f0 direction alone sufficed for males. Accuracy rates reached 84–87% for females and 95–97% for males, highlighting the potential of MLAs in analyzing tonal languages and suggesting broader applications across various fields.
Publisher
Humanities and Social Sciences Communications
Published On
May 10, 2024
Authors
Amalesh Gope, Anusuya Pal, Sekholu Tetseo, Tulika Gogoi, Joanna J, Dinkur Borah
Tags
machine learning
tonal contrasts
Chokri language
acoustic correlates
supervised algorithms
accuracy rates
fundamental frequency
Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny