基于改进粒子群算法和支持向量结合的辅助发声训练系统研究

Research on an auxiliary voice training system based on improved particle swarm optimization algorithm and support vector

随着技术进步,尤其是在人工智能领域,全球交流日益紧密,使得跨语言的人机交互技术成为生活中的重要部分。此次研究针对英语发音的问题以及与其相关的设备控制难题,开发了一种结合了优化粒子群算法和支持向量回归的辅助发音训练系统。该系统首先对粒子群算法进行了创新,融入了混沌机制。然后,该机制与支持向量回归算法相结合。研究结果表明,与传统方法相比,这种新的结合策略大大减少了计算次数和成本,标准粒子群算法的平均计算次数是23次,混沌版本为17次,而遗传版本为13次,所提方法只需7次。由此可见,这种新方法在搜索能力和准确性上均有所增强,为实际应用带来了巨大的价值。

With the advancement of technology,especially in the field of artificial intelligence,global communication is becoming increasingly close,making human-computer interaction technology across languages an important part of life.In order to solve the problem of English pronunciation and its related equipment control problems,an auxiliary pronunciation training system combining optimized particle swarm optimization algorithm and support vector regression is developed.Firstly,the system innovates particle swarm optimization algorithm and integrates chaos mechanism.This mechanism is then combined with the support vector regression algorithm.The results show that compared with the traditional method,the new combination strategy greatly reduces the computation times and cost.The average computation times of the standard particle swarm algorithm is 23 times,the chaotic version is 17 times,and the genetic version is 13 times,and the proposed method only needs 7 times.It can be seen that this new method has improved the search ability and accuracy,and has brought great value for practical application.