CAJ | 학술논문

文章在分析语音合成、语音转换原理和实现方法的基础上，将这两种技术相结合，设计实现了一种具有更多人类个性特征的语音合成转换系统。完成了基于Microso在SAPI5．1的文语转换，设计实现了基于高斯混合模型GMM和残差预测的语音转换系统，并对该系统进行了主客观评价；理论分析和实验结果表明’了本文设计系统的有效性。
문장재분석어음합성、어음전환원리화실현방법적기출상，장저량충기술상결합，설계실현료일충구유경다인류개성특정적어음합성전환계통。완성료기우Microso재SAPI5．1적문어전환，설계실현료기우고사혼합모형GMM화잔차예측적어음전환계통，병대해계통진행료주객관평개；이론분석화실험결과표명’료본문설계계통적유효성。
This paper has fulfilled a conversion system of speech synthesis with more individual characteristics, combining with the technologies of speech synthesis and voice conversion. Text-to-speech conversion is completed based on Microsoft SAPI 5.1, then a voice conversion system is designed based on gaussian mixture model GMM and residual prediction. And this system performance are evaluated using subjective and objective methods; Theoretical analysis and experimental results show that the system is more effective.