|
|
|
Information Access by Voice |
|
Imagine a world where being connected did not require education, training and expensive
technology. Where it did not require literacy and a knowledge of English. Where it was
possible for everyone to use a simple, widely available device
to access information, use email and even carry out on-line transactions
in their own language. And where it was as easy to develop and maintain those
on-line services as it is today to create standard web services.
| | | |
Such a world is more than possible. The telephone, mobile or landline, has become
the ubiquitous connecting device throughout the world. Over the last few years, the
VoiceXML forum has defined & refined a way of authoring voice-based services. All that
is missing now is the speech technology that underpins such services - text to speech
(TTS) to deliver information, and automatic speech recognition (ASR) to access
it and control its delivery.
| | | |
But here's the issue - both are complex technologies that require both
linguistic and specific technical expertise to develop, and are mostly only
available for the world’s major languages, and at considerable cost.
| | |
Computer Use for Visually Impaired |
|
There are various IT products available world-wide which assist the visually impaired to use IT tools for word processing or for surfing the internet. One highly effective solution is a screenreader, which speaks aloud what is on the computer screen. However, as for the voice services described above, the TTS technology which underpins screenreaders is mostly only available for the world’s major languages. Ironically, even in countries where one of the major languages are taught in schools, visually impaired people have least access to the school system and therefore tend to know only their local languages.
|
| | LLSTI |
|
The Local
Language Speech Technology Initiative (LLSTI) has been solving the technology gap by
bringing together motivated groups in developing countries, providing tools, expertise,
support and training to enable first TTS and then ASR to be developed in their
own local languages. The systems developed have been made freely available under an
open source license, with the aim of building a community of interest to
further improve the systems and minimize the barrier to their actual
deployment. LLSTI partners also work in partnership with organisations
interested in deploying systems using the technology.
| | |
|
Top
|
|
