Cargando…

Influence of Diversified Health Elements Based on Machine Learning Technology on Pop Vocal Singing in a Cultural Fusion Environment

The multicultural environment is affected by the ongoing advancement of science and technology, which results in more and more planned cultural fusions and collisions between various cultures. The emergence of distinct national cultures has emphasised cultural diversity. Music naturally takes the in...

Descripción completa

Detalles Bibliográficos
Autor principal: Xia, Han
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Hindawi 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9529412/
https://www.ncbi.nlm.nih.gov/pubmed/36200080
http://dx.doi.org/10.1155/2022/7903838
Descripción
Sumario:The multicultural environment is affected by the ongoing advancement of science and technology, which results in more and more planned cultural fusions and collisions between various cultures. The emergence of distinct national cultures has emphasised cultural diversity. Music naturally takes the initiative and promotes diversity in social and cultural awareness as a cultural art form with distinctive charm. Cultural variables play a significant role in the development, appeal, and wide transmission of voice output. It is an authentic catharsis and a vivid record of spiritual activity among people. Because the diversity of art is also the source of the legacy and growth of creative innovation, the diversified integration of art will also promote the shared development of all nations. Vocal music and singing art must adapt to the circumstances, follow the trend of the times, and grow slowly and healthily in the direction of diversity in the context of multicultural development. Musical emotion is the key component of music. The periodic properties of sound should be studied since they have important implications for music study. In order to learn and predict the 8-dimensional emotion vector of musical compositions, this study creates a dataset of 200 pieces of music, isolates music emotion detection as a regression issue, and applies machine learning techniques. According to experimental findings, when mid- and high-level characteristics are used as input instead of low-level features, accuracy can increase from 50.28% to 68.39%.