English Language Synthesis

1221 Words5 Pages
ABSTRACT Text to Speech Conversion System is basically used for converting text into equivalent speech. Research is going on to produce more natural speech. Lot of work is already done in developing English Text to Speech system. In this paper we are explaining method to perform transliteration from Hindi to English and vice versa and simple approach to improve voice quality Hindi Text to Speech System. We are considering two types of input. In first method User can manually enter any text in Hindi language by using English characters (i.e. using transliteration). In Second method user can browse any text file written in Hindi (Devnagari Script). Single .mp3 sound file will be generated as an output which represents speech equivalent to given…show more content…
Existing text to speech technology produces speech but quality of speech is not so natural. It is much like robotic speech. Research is going on in developing natural, fluent speech. There is still a long way to reach the goal. Compared to Indian Languages more work is done for foreign languages related to text to speech conversion. However researcher are now concentrating on certain areas like prosodic, text preprocessing and pronunciation in order to produce natural and pleasant speech, improve voice quality and linguistic analysis. Hindi Language is used by more than 400 million people across world .As there is need to research in text to speech synthesis for Hindi Language, we are proposing a solution to improve the quality of Text to Speech System so that it will produce more fluent, natural Hindi speech sound. This can be helpful for people who cannot read Hindi language, vision disable person…show more content…
Ramakrishnan, Partha Pratim Talukdar addresses the problem of Hindi compound word splitting and its relevance to developing a good quality phonetizer for Hindi Speech Synthesis. The constituents of a Hindi compound word are not separated by space or hyphen. Hence, most of the existing compound splitting algorithms cannot be applied to Hindi. They propose a new technique for automatic extraction of compound words from Hindi corpus [2]. N. Sridhar Krishna, Partha Pratim Talukdar, Kalika Bali, A.G. Rama Krishnan reports preliminary results of data-driven modeling of segmental (phoneme) duration for Hindi .They have presented Classification and Regression Tree (CART) based data-driven duration modeling for segmental duration prediction[3] . Lakshmi Sahu and Avinash Dhole explained single text-to-speech (TTS) system for Indian languages (Viz., Hindi, Telugu, Kannada etc.) to generate human voice or speech (text to a spoken waveform). In a text-to-speech system, spoken utterances are automatically generated from text. They present a corpus-driven text-to-speech (TTS) system based on the concatenative synthesis approach

More about English Language Synthesis

Open Document