| Autor: |
Manideep, Nettyam, Mohana, J. |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2023, Vol. 2821 Issue 1, p1-7, 7p |
| Abstrakt: |
The objective of this research is to design a Novel Statistical Gaussian Mixture (GMM) Machine learning model for recognizing the voice. The accuracy of recognition rate is compared with the Hidden Markov model (HMM). The research includes two groups, in which Statistical Gaussian Mixture is group 1 and Hidden Markov Model is group 2. Github Dataset is used for collecting different voices. The total sample size is 40 and the sample size for eac h group is 20 and the sample size was calculated using G power (80%) and alpha (0.05). The novel Statistical Gaussian Mixture model has achieved 96.2695% of accuracy and 0.59649 standard deviation while Hidden Markov model has achieved 86.2895% of accuracy and 0.62618 standard deviation. The significant accuracy value of the algorithms is 0.000 (p < 0.05, 2 tailed). The performance of recognizing voice using Statistical Gaussian Mixture model has provided better accuracy than Hidden Markov Model. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
