20 Conversion of English Text-To-Speech (TTS) Using Indian Speech Signal

The objective of this paper is to convert the english text into speech. The conversion of english text into speech is done by using a stored speech signal data. Text to speech conversion module is designed by the use of matlab. By the use of microphone the phonemes (alphabets, numbers, words) are recorded using a goldwave software. The recorded .wav (sounds) files are saved as a database separately. The phonemes are extracted from the text file. For text to speech conversion the concatenation method is proposed. The recorded speech are concatenated together to produce the synthesized speech. The resulting speech output is assessed by listening test. The Mean Opinion Score (MOS) value is calculated for the synthesized speech output as the performance measure. In future work, a miniaturized hardware implementation will be developed for helping visually impaired persons in understanding text they come across in day today life.