Introduction - If you have any usage issues, please Google them yourself
This method is so popular lies in its inherent statistical framework: the training the limited voice training set data model approximation parameter is simple Model according to the special vocabulary, voice changes such as size, type or model of cognitive system architecture of flexible Easy and convenient to realize the whole cognitive system. In this article on the explanatory, we will explain the application is very important in speech recognition of statistical method, and discussed a series of has not solve the problem of the rational and practical, because they are important and has a great influence on the different performance of the system implementation.
。