基于音素发生率的自动语言辨识

Automatic Language Identification Using the Frequencies of Occurrence of Phones

不同语言的语音基元的种类和数量存在着差异,即使两种语言有相同的音素,它们的发生频率也存在差异。以前基于音素标识的语言辨识系统,难以引入新的语言。本文分别使用了GMM和VQ模型对音素符号发生率信息在语言辨识中的作用进行了研究,使用了音素符号发生率方法以及三种改进方法,各项实验结果表明音素符号发生率信息在语言辨识中具有一定的作用,可以作为语言辨识方法研究的一个方向。

Phonetic inventories differ from language to language. Even when languages have identical phones, the frequencies of occurrence of phones differ across languages. It＇ s difficult to introduce new languages when the language identification system used phones label. In this paper, we study the frequencies of occurrence of phones using Gaussian Mixture Model and Vector Quantization. The method of occurring of phones and three improved methods are provided in this paper. The experimental results show the frequencies of occurrence of phones are very effective in language identification.