Accurate Analysis of Voice Disorder Using ResNet-50 Algorithm in Comparison with ResNet-18 Algorithm

Aakash S., Bharatha N.

2023

Abstract

The study aims to enhance voice disorder detection precision using the novel RESNET-50 algorithm and comparing its efficacy with the resnet-18 algorithm. for evaluating the accuracy of voice disorder identification, the research uses a confidence level of 95% and a g power of 0.8. two algorithms, novel RESNET-50 and RESNET-18, are applied to a dataset of 864,448 mp3 audio files with accompanying metadata. the findings reveal that the novel RESNET-50 algorithm boasts an accuracy of 88.70%, superior to the 70.81% achieved by the resnet-18 algorithm. however, with a significance value of 0.18 (independent sample t-test p<0.05), no noteworthy statistical difference was found between the two. in essence, the novel RESNET-50 algorithm demonstrates a higher accuracy in voice disorder analysis compared to the RESNET-18 algorithm.

Paper Citation

in Harvard Style

S. A. and N. B. (2023). Accurate Analysis of Voice Disorder Using ResNet-50 Algorithm in Comparison with ResNet-18 Algorithm. In Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT; ISBN 978-989-758-661-3, SciTePress, pages 502-507. DOI: 10.5220/0012564600003739

in Bibtex Style

@conference{ai4iot23,
author={Aakash S. and Bharatha N.},
title={Accurate Analysis of Voice Disorder Using ResNet-50 Algorithm in Comparison with ResNet-18 Algorithm},
booktitle={Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT},
year={2023},
pages={502-507},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012564600003739},
isbn={978-989-758-661-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT
TI - Accurate Analysis of Voice Disorder Using ResNet-50 Algorithm in Comparison with ResNet-18 Algorithm
SN - 978-989-758-661-3
AU - S. A.
AU - N. B.
PY - 2023
SP - 502
EP - 507
DO - 10.5220/0012564600003739
PB - SciTePress