基于RBF径向基网络的演唱评价模型研究

摘要

针对演唱评分方法易受到主观因素的影响,提出采用广义回归RBF神经网络模拟评委进行稳定打分并分析网络结构在声乐评价中的优势,对声乐艺术评价建立评价指.将定性评价转换为定量成绩向量输入神经网络生成评分.用单声部民族唱法进行实验,将输出结果与预期分值进行比对.结果表明,该模型能有效降低人为因素的干扰,模拟专家进行稳定打分.%To giving a correct assessment of a singer's performance is affected by multiple factors. GRNN RBF neural network can simulate a nonlinear system. This article establishes the evaluation model and use neural network to output a reasonable score. At last,this article uses an example to analysis the feasibility of the model. The experimental results show that the model can effec?tively reduce the interference by human in folk singing evaluation.