利用小词汇量混合语音识别系统中的HMM状态所形成的规则的二维阵列，对状态观测概率进行分解。
By using the regular two-dimensional array of HMM states in a hybrid speech recognition system with small vocabulary size, the factorization of observation probabilities is performed.
这就要求对词汇的意义进行完整的分析，并尽可能地将之分解成为最小元素，从而使语言的语义能够从形式上得到计算或组合。
It is required that the meanings of lexicon be fully analyzed and, if possible, be decomposed in nearly primitive, so that the semantics of language can be computable in a formal way.
经过批准的详细项目范围说明书，相应的工作分解结构，以及工作分解结构词汇表是项目的范围基准。 ！
The approved detailed project scope statement and its associated WBS and WBS dictionary are the scope baseline for the project.
应用推荐