Affiliation:
1. School of Art , Xichang University , Xichang , Sichuan , , China .
Abstract
Abstract
In this paper, the extracted acoustic features are processed using the MFCC method to classify the sound recognition. The linear spectrum of the sound is mapped onto the Mel FPP coefficients and Mel nonlinear spectrum, respectively, and the MFCC is obtained by cepstrum processing. The Mel FPP coefficients and Gaussian mixture model are combined to create the acoustic model. On this basis, the art and emotional skills of vocal singing are explored, and simulation and empirical experiments are set up to analyze the effect of the model constructed in this paper. The experimental results show that the model constructed in this paper tends to stabilize when the number of iterations is 2, the final accuracy rate stabilizes near 0.9, and the model is effective. Different vocal types are recognized by the model constructed in this paper, and the recognition rate of emotional expression of all vocal types is above 0.7, and the model recognition rate is high. The accuracy and recall classification for emotional expression techniques are 0.73 and 0.79, resulting in a more balanced evaluation overall.
Reference15 articles.
1. Chen, Wen, Pang, Guofei, Holm, & Sverre, et al. (2017). Fractional biharmonic operator equation model for arbitrary frequency-dependent scattering attenuation in acoustic wave propagation. The Journal of the Acoustical Society of America.
2. Kohlrausch, Armin, Vecchi, Alejandro, Osses, & Lachenmayr, et al. (2017). Predicting the perceived reverberation in different room acoustic environments using a binaural auditory model. The Journal of the Acoustical Society of America.
3. Sun, J. (2020). Research on resource allocation of vocal music teaching system based on mobile edge computing. Computer Communications, 160(2).
4. Chi, X. (2017). Study on vocal music teaching innovation mode based on computer simulation and voice spectrogram analysis. Revista de la Facultad de Ingenieria, 32(16), 400-406.
5. Murthy, Y. V. S., & Koolagudi, S. G. (2018). Classification of vocal and non-vocal segments in audio clips using genetic algorithm based feature selection (gafs). Expert Systems with Applications, 106(sep.), 77-91.