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

<p>Opera singing is an indispensable part of college voice students’ study, and in response to the problem of insufficient opera singing ability of college students, this paper proposes a method of applying vocal feature extraction technology to the teaching of opera singing. The method takes the vocal frequency frames in the students’ opera singing as the unit and uses the short-time energy to datamaterialize the vocal frequency features of the students’ western opera singing, which reflects more clearly the information of the vocal frequency signals of western opera singing in the time domain and frequency domain. The final feature vector is formed by linearly splicing all vocal features together using Multi-Feature Linear, which is combined with a classifier to categorize different vocal features. By analyzing the correlation between students’ short-time ability and music tempo in Western opera singing, students’ music control ability was improved, and by visualizing the music pitch data, students were able to have a clear perception of the difference between their pronunciation and that of the original voice, thus improving their pitch. The results show that the correct rate of students’ onset recognition in the 20 Western opera samples is above 0.98, among which the onset recognition rate of sample 18 is 0.99803, indicating that students have better rhythmic control of Western opera singing. Practicing according to their pronunciation deficiencies after vocal feature extraction resulted in different degrees of increase in singing scores for the students.</p>
</abstract>

Opera singing is an indispensable part of college voice students’ study, and in response to the problem of insufficient opera singing ability of college students, this paper proposes a method of applying vocal feature extraction technology to the teaching of opera singing. The method takes the vocal frequency frames in the students’ opera singing as the unit and uses the short-time energy to datamaterialize the vocal frequency features of the students’ western opera singing, which reflects more clearly the information of the vocal frequency signals of western opera singing in the time domain and frequency domain. The final feature vector is formed by linearly splicing all vocal features together using Multi-Feature Linear, which is combined with a classifier to categorize different vocal features. By analyzing the correlation between students’ short-time ability and music tempo in Western opera singing, students’ music control ability was improved, and by visualizing the music pitch data, students were able to have a clear perception of the difference between their pronunciation and that of the original voice, thus improving their pitch. The results show that the correct rate of students’ onset recognition in the 20 Western opera samples is above 0.98, among which the onset recognition rate of sample 18 is 0.99803, indicating that students have better rhythmic control of Western opera singing. Practicing according to their pronunciation deficiencies after vocal feature extraction resulted in different degrees of increase in singing scores for the students.

Research on the Cultivation of Western Opera Singing Ability of College Vocal Music Students Based on Multi-feature Fusion Technology

Chugye University for the Arts, Bukahyeon-dong

Applied Mathematics and Nonlinear Sciences

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

Opera singing is an indispensable part of college voice students’ study, and in response to the problem of insufficient opera singing ability of college...

Research on the Cultivation of Western Opera Singing Ability of College Vocal Music Students Based on Multi-feature Fusion Technology

Data publikacji: 31 sty 2024

Zakres stron: -

Otrzymano: 02 sty 2024

Przyjęty: 08 sty 2024

DOI: https://doi.org/10.2478/amns-2024-0313

Słowa kluczowe
Acoustic feature extraction, Short-time energy, Vocal frequency signal, Western opera singing

© 2024 Mingzhu Chen, published by Sciendo

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

Research on the Cultivation of Western Opera Singing Ability of College Vocal Music Students Based on Multi-feature Fusion Technology

Data publikacji: 31 sty 2024

Zakres stron: -

Otrzymano: 02 sty 2024

Przyjęty: 08 sty 2024

DOI: https://doi.org/10.2478/amns-2024-0313

Słowa kluczoweAcoustic feature extraction, Short-time energy, Vocal frequency signal, Western opera singing

© 2024 Mingzhu Chen, published by Sciendo

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

Słowa kluczowe
Acoustic feature extraction, Short-time energy, Vocal frequency signal, Western opera singing