论文部分内容阅读
在语音合成技术的研究中,情感语音合成是当前研究的热点.在众多研究因素中,建立恰当的韵律模型和选取好的韵律参数是研究的关键,它们描述的正确与否,直接影响到情感语音合成的输出效果.为了攻克提高情感语音自然度这一难点,对影响情感语音合成技术韵律参数进行了分析,建立了基于关联规则的情感语音韵律基频模型.本文通过研究关联规则、改进数据挖掘Apriori算法并由此来获得韵律参数中基频变化规则,并为情感语音合成的选音提供指导和帮助.
In the research of speech synthesis, emotional speech synthesis is the hotspot of current research.In many research factors, to establish the appropriate prosodic model and select the prosodic parameters is the key to the study, they describe the correct or not, directly affect the emotion In order to overcome the difficulty of improving the naturalness of emotional voice, the prosodic parameters affecting emotional speech synthesis are analyzed and the basic phonetic frequency model based on association rules is established.Through the study of association rules, improvement of data Mining Apriori algorithm and thus to obtain the rhythm parameters of the fundamental frequency variation rules, and for emotional speech synthesis to provide guidance and help.