汉语语音的时域声调转换方法

The Time-Domain Transformation of Chinese Speech Tone

简要地介绍了用于语音分析合成的时城基音同步叠加算法，在此基础上提出一种汉语语音时域声调转换方法。利用这种方法可以将一种声调的语音转换为另一种声调的语音，除微小的音质降低外，仍可保持较好的语音质量。这种方法直接对语音波形进行处理，具有计算简单、能在一般微型计算机上进行实时的特点。将之用于语音合成系统，可以通过相同声韵母的音节只存储一种声调的语音数据而大大降低音库的容量；用这种方法按照汉语语句的语调变化规律来合成语句，还可以较好地改善汉语语句合成的自然度．

Introduces the time-domain pitch-synchrDnous overlap-add algorithm for speech analysissynthesis. Based on this, a time-domain iran formation method of Chinese speech tone is presented. The method is capable of transforming a speech waveform from one tone into another, and keeping good quality of sound only with a little decline in sound quality.The method is directly applied to the speech waveform, and therefore contains the features of simplicity in computation and being able to procees speech signal in real-time on ordinary personal computers. By using the method, the required capacity of speech database for speech synthesis can be reduced, and sentence naturalness can be improved by modifing the intonation of output sentences.