一种基于决策树模型的音库构建和基元选取方法

针对嵌入式设备的存储容量小、计算能力有限的特点，设计了一种基于CART（Classification and Regression Trees）决策树模型的基元预选算法和基元选取算法，可以从原始语音语料库中挑选出最有代表性的基元样本，从而有效地降低音库规模和算法的复杂度，满足了嵌入式TFS（Text-to-Speech）系统的需要。基于以上算法，移动终端上实现了一个嵌入式中文TTS系统，实验结果表明该系统的合成语音具有较高的可懂度和自然度。

A Decision Tree Based Approach for Construction of Speech Database and Unit Selection

A CART based unit pre selection algorithm, which is to select the most representative units from the original speech corpus to reduce the footprint of the units inventory, and a unit selection algorithm with low complexity are described in this paper. Based on these two algorithms, an embedded Chinese TTS system is implemented on mobile device. The experiments show that this system can produce speech with high intelligibility and naturalness.