Text-to-Speech Synthesis using Phoneme Concatenation

Text-to-speech (TTS) synthesis transforms any linguistic information stored as data or text into speech. It is widely used in audio reading devices for blind people. In the last few years however, the use of TTS technology has grown far beyond the disabled community and become a major adjunct to the rapidly growing use of digital voice storage for voice mail and voice response systems. Concatenative TTS synthesis system has gained in popularity in recent years, due to its more natural sounding synthesized speech. It concatenates pre-recorded speech units into the word sequences according to the pronunciation dictionary or set of rules [i]. For general purpose TTS, it must be able to read unrestricted text [ii]. Thus it is desirable to have the basic speech units much smaller, like for example phonemes or diaphones, in order to be able to synthesize all possible phonetic and prosodic variation in the language with a limited database size.