基于蚁群算法的语音信号动态时间规划

蚁群算法是一种新型的随机优化算法,应用蚁群算法优化机制,提出了一种基于蚁群算法的语音信号动态时间规划方法———蚁群动态时间规划算法,搜索语音信号之间匹配的一条全局最优路径,进而以此衡量语音信号之间的相似度.算法给出了蚁群状态转移概率及信息素更新方程,既利用了语音信号的全局特征又考虑了其局部信息.理论分析与仿真实验结果均证明了此方法的可行性,与传统的DTW算法相比较,其匹配结果更能体现匹配语音信号之间的相似度.

Speech sgnal dynamic time programming based on ant colony algorithm

Ant Colony Algorithm is a novel random optimization algorithm. It has shown many promising properties in solving complicated optimization problem. Applying the thought of ant colony algorithm to speech signal processing is presented a new dynamic time programming based on ant colony algorithm Time the si Programming. The algorithm is used to search a global optimization path which is indicated milarity between the speech signals. The new state transfer probability and updating rule Ant Dynamic the degree of of the pheromone intensity use both the global and the local character of speech signal. Theoretic analyses and simulation experiment all certify the method feasibility. The matching result of the new method shows more accurate similarity between speech signals than the traditional DTW method in the experiments.