我们可以通过在有向图$G=(V,E)$中使用动态规划的算法来实现语音识别功能。对每条边$(u,v) in E$打上一个声音标签$sigma (u,v)$,该声音来自于有限声音集$sum$ 。图中从特定的顶点$v_0 in V$开始的每条路径都对应模型产生一个可能的声音序列。对于一条有向路径,我们定义标签为路径中边的标签的简单连结。
Biological applications often need to compare the DNA of two (or more) different organisms. A strand of DNA consists of a string of molecules called bases, where the possible bases are adenine, guanine, cytosine, and thymine. Representing each of these bases by its initial letter, we can express ...