The Study of Various Emotionally-sounding Classification using KNN, Bayesian, Neural Network Methods

Autor: Saman Rajebi, Sina Andarabi, Alireza Nobakht
Rok vydání: 2020
Předmět:
Zdroj: 2020 International Conference on Electrical, Communication, and Computer Engineering (ICECCE).
DOI: 10.1109/icecce49384.2020.9179451
Popis: The human voice comes through the vocal cords. The voice can be divided into the lung, the vocal cords, and the larynx. Human emotions can be detected through the voice, and because of that, it is important to process the human voice. Sadness and happiness, as well as truthfulness and deceitfulness, can be detected through the voice. The use of these methods could reduce crime and etc. Data recorded from a variety of actors were used to detect emotions portrayed by human voices. The k nearest neighbor (KNN), Bayesian, and radial basis function (RBF) methods were used to detect emotions by training and testing algorithms based on the data. The KNN method had a low correct classification rate (CCR), and the RBF method had a high CCR.
Databáze: OpenAIRE