论文部分内容阅读
本文提出了与传统的单一单数的分类方法截然不同的多参数，多层次的分类方法，与单一参数的分类方法比较而言，多参数、多层次的分类方法能很好地适应单音节因上下文而产生的变化，具有鲁棒性好、适应性强的特点，能很好地对自然语音流单音节这种变化性很强的对象进行分类．
In this paper, we propose a multi-parameter and multi-level classification method which is very different from the traditional single singular classification method. Compared with the single parameter classification method, the multi-parameter and multi-level classification methods can well adapt to the monosyllabic context The resulting changes, with robustness and adaptability, are good at classifying highly variable objects as natural monosyllabic syllables.