Reading without Seeing: Mahiti's work on eSpeak Aug. 27, 2013, 9:32 a.m.

Reading without Seeing: Mahiti's work on eSpeak

Summary :

Espeak - Text to Speech Software (Open Source)– Enabling the visually impaired to access information through technology

About Text to Speech Software (TTS)

People who are literate and can see, are blessed with the ability to read. To them information of any kind, be it political, social, or even about entertainment are easily available through various forms of media. Those who are challenged with visual impairment are not so lucky to be able to read a newspaper, magazine or a blog post. They have to use their hearing ability to read, understand and comprehend. This is exactly where Text to Speech software(TTS) is of great importance. The software enables them to feel equal to the rest of the people as far as access to information is concerned.

A number of proprietory TTS programmes are available now days which are, to most of the visually impaired people, difficult to afford or have technical obstacles. eSpeakis a compact open source software speech synthesizer.

It uses a "formant synthesis" method which allows many languages to be provided in a small size. The speech is clear, and can be used at high speed, but is not as natural or smooth as larger synthesizers which are based on human speech recordings.

Most of the work does not need any programming knowledge. Just an understanding of the language, awareness about its features, patience and attention to detail.Wikipediais a good source of basic phonetic information.

In many cases it should be fairly easy to add a rough implementation of a new language, hopefully managable enough to be intelligible. After that it is a gradual process of improvement.

It is possible to have different variants of a language for different dialects. For example the sound of some phonemes are changed, or some of the pronunciation rules differ.

My thoughts With all the limitations about ambiguous pronunciation rules of indic languages and intonation issues, most of the visually impaired people I met across India, prefer eSpeak to all the other TTS solutions. We are trying to include and improvise as many Indic languages as possible, into eSpeak.

I could add Bengali in eSpeak version 1.47.6. We are currently in the process of adding a few North East Indian languages too. I will share my experience on adding bengali to eSpeak soon.