Modal Analysis System of Emotion First in Vocal Art Under the Background of Sound Wave Recognition and Matching

This paper focuses on the back-end pattern matching algorithm, and introduces the fuzzy-mean clustering algorithm in the model training process, which makes the recognition effect significantly improved. This algorithm has high robustness in clustering. It is introduced into the training algorithm to replace the commonly used hard clustering algorithm to find the initial point of the model. The importance of emotional expression in vocal performance art is analyzed, and then it is discussed. The measures to improve the emotional expression level of vocal performance art hope to bring help to the majority of colleagues to improve their artistic level. Correspondingly starts n sub-threads for concurrent computing. When the improved feature coefficients are applied to the Android platform, the equivalence rate is reduced by 7.24%, the recognition rate is increased by 6.8%, and the robustness in noisy environments has also been significantly improved. After adding 3 sub-threads, the computation time of feature extraction is shortened by about 35%.