Speech Synthesis Technology And Mobile Voice Dictionary

With the development of mobile phone technology and popularization of its user, mobile phone software is more and more abundance. Mobile phone software is belongs to embedded software, it should use little space and run efficiency. Electronic dictionary and speech synthesis are all running well in PCs, but it is difficult to be embedded in mobile phones, it should be special designed to adapt mobile phone. This thesis deeply researches on speech synthesis and data compress theory, and successfully complete speech dictionary on mobile phone. The main researches are follows:1. This thesis researches on speech synthesis theory, articulatory synthesis technology, LPC synthesis and formant synthesis technology.2. Now waveform coding synthesis technology is one of the most important speech synthesis method, it can produce natural and articulate speech. Pitch-Synchronous Overlap adding (PSOLA) method can easily adjust pitch' s rhythm and base frequency. This thesis researches on PSOLA' s principle and compare different PSOLA method, and researches on pitch estimation algorithm and base frequency estimate arithmetic.3. The data of English-Chinese and Chinese-English dictionary is mass. For mobile phone it should design high efficiency data compress method. This thesis researches on data compress method, firstly give out high efficiency data compress method for English Chinese dictionary. Using Huffman code to compress English word and phonetic symbol, using"Statistic replacement compress method" to compress Chinese word and dictionary symbol. The compression ratio is 56%, and accomplish English Chinese dictionary on mobile phone. Compare to other mobile phone dictionary, it uses little space of ROM and RAM, and the speed is faster.4. Researches on speech synthesis on mobile phone, Using diphone and PSOLA method, successfully complete English synthesis on mobile in 2M space ROM, the quality of English speech synthesis is fine.