A rule-driven algorithm based on an iterated finite state transducer is proposed in present research to automatically convert graphemes to corresponding phonemes.

• 可以使用基于规则驱动的迭代有限状态转录机来解决这一问题。