上QQ阅读APP看书,第一时间看更新
Preface
Using Hidden Markov Models (HMMs) is a technique for modeling Markov processes with unobserved states. They are a special case of Dynamic Bayesian Networks (DBNs) but have been found to perform well in a wide range of problems. One of the areas where HMMs are used a lot is speech recognition because HMMs are able to provide a very natural way to model speech data. This book starts by introducing the theoretical aspects of HMMs from the basics of probability theory, and then talks about the different applications of HMMs.