Text to Phoneme Conversion in Persian Using Neural Networks

Speech is the most natural and widespread from of human communication. That’s why speech synthesis has interested researchers for decades. In this paper, a Persian text to speech system is presented. The system uses speech waveform concatenation method that is comparatively mature in text – to – speech synthesis. This paper discusses the experimental study on the use of neural network in text – to – speech systems for Persian language. In the context of text to phoneme conversion, the neural network demonstrate good performance. It is shown that a network can capture significant portion of regularities in the Persian pronunciation as well as absorb many of the irregularities .