汉语文语转换系统中可训练韵律模型的研究

针对汉语的韵律特征受语境参数影响时,表现出层次性的特点,本文描述了一种带特殊加权因子和输出优化功能的人工神经网络,并用其来构筑汉语TTS系统的韵律模型。大量测试表明,该人工神经网络的拓扑结构相较传统的人工神经网络模型更能反映出汉语的韵律特点。它提高了模型本身的收敛速度和运算精度,从而改善了整个韵律模型的质量。同时,本文还对汉语音节的基频曲线进行了规格化处理,较详细的分析了音节基频规格化参数-SPiS,在基频调节中的作用和方式。SPiS参数能够反映出汉语的声调特点,且方便了网络模型的建立和汉语韵律的控制。

The study of the trainable prosodic model for Chinese text to speech system

Mandarin prosody is characterized by its hierarchical structures when it is influenced by the context. An artificial on this, a neural network, with specially weighted factors and optimizing outputs, is described and applied to construct the Mandarin prosodic model in a TTS system for Chinese. Extensive tests show that the structure of the artificial neural network characterizes the Mandarin prosody more exactly than traditional models. Learning rate is speeded up and computational precision is improved, which makes the whole prosodic model more efficient. Furthermore, the paper also stylizes the Mandarin syllable pitch contours with SPiS parameters (Syllable Pitch Stylized Parameters), and analyzes them in adjusting the syllable pitch. It shows that the SPiS parameters effectively characterize the Mandarin syllable pitch contours, and facilitate the establishment of the network model and the prosodic controlling.