<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>We must seize the chance for growth in the big data era for innovation and optimization to efficiently optimize music education instruction in colleges and universities. This paper applies the SVM method to studying vocal recognition in popular music to build a music learning aid system. The system gives the exact location of the intro, bridge, and outro through the recognition of vocals in pop music, which helps learners to practice in a targeted way. MFCC features are used, and low-pass filtering is applied to the classification results later, resulting in a recognition rate of 85.74% on a frame-by-frame basis. From the test results, SVM has a better generalization ability than other classifiers, including ANN, GMM, and HMM, with a recognition rate of at least 8.84% higher. In the practicality test, the experimental class applied the music-assisted system for learning, and the comparison class used the traditional teaching method. 46 more people in the experimental class rated good music ability than the control class, 68 more rated good innovation status than the control class, and 71 more rated good independent learning ability than the control class. Therefore, big data technology must be used to innovate music education in colleges and universities to increase teaching efficacy successfully.</p>
</abstract>

We must seize the chance for growth in the big data era for innovation and optimization to efficiently optimize music education instruction in colleges and universities. This paper applies the SVM method to studying vocal recognition in popular music to build a music learning aid system. The system gives the exact location of the intro, bridge, and outro through the recognition of vocals in pop music, which helps learners to practice in a targeted way. MFCC features are used, and low-pass filtering is applied to the classification results later, resulting in a recognition rate of 85.74% on a frame-by-frame basis. From the test results, SVM has a better generalization ability than other classifiers, including ANN, GMM, and HMM, with a recognition rate of at least 8.84% higher. In the practicality test, the experimental class applied the music-assisted system for learning, and the comparison class used the traditional teaching method. 46 more people in the experimental class rated good music ability than the control class, 68 more rated good innovation status than the control class, and 71 more rated good independent learning ability than the control class. Therefore, big data technology must be used to innovate music education in colleges and universities to increase teaching efficacy successfully.

The rise of big data on the development of music education and innovation to promote

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

We must seize the chance for growth in the big data era for innovation and optimization to efficiently optimize music education instruction in colleges and...

The rise of big data on the development of music education and innovation to promote

Data publikacji: 09 wrz 2023

Zakres stron: -

Otrzymano: 11 paź 2022

Przyjęty: 20 mar 2022

DOI: https://doi.org/10.2478/amns.2023.2.00321

Słowa kluczowe
Big data, Music education, SVM, Human voice recognition, MFCC features

© 2023 Tie Wang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

The rise of big data on the development of music education and innovation to promote

Data publikacji: 09 wrz 2023

Zakres stron: -

Otrzymano: 11 paź 2022

Przyjęty: 20 mar 2022

DOI: https://doi.org/10.2478/amns.2023.2.00321

Słowa kluczoweBig data, Music education, SVM, Human voice recognition, MFCC features

© 2023 Tie Wang, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Słowa kluczowe
Big data, Music education, SVM, Human voice recognition, MFCC features