We propose to analyze and model the phonetic and acoustic confusuons separately,using pronunciation dictionary and acoustic model merging technology to create a new model,with a significant 15% absolute WER reduction on the Chinese English,which only sacrifics 1% recognition rate on the native English.

There are many homophones and polyphones in Chinese personal names and geographical names,resulting in some defect in the database query technology.

In Text-to-Speech(TTS) systems,Grapheme-Phoneme(G2P) conversion is one of the most important modules,and polyphone disambiguation is its key problem.

In this paper, a new rule-based method of automatic phonetic notation on the 220 polyphones whose appearing frequency exceeds 99% is proposed.

The existence of Chinese homonym and polyphony words gives our country the increased difficulty for applications of computer.

The article includes 3 parts: Part One introduces a solution to polyphony words discretion and processing in Chinese system; Part Two probes a method of Chinese summary generating and processing; Part Three discusses and explains the problems in automatic word segmentation.

This paper systematizes in detail the homophones of Haikou dialect, pointing out the main phonetic differences between the old-style and new-style Haikou dialect, and attempting to minor the complexity of polyphones in Haikou dialec