Home  Areas of research
|
The Lwazi telephone-based, speech-driven information system will allow easy access to government information and services to all South Africans and will showcase the potential of human language technologies (HLTs) in South Africa. The three main areas of research for the Lwazi project are:
- Application selection and human factors
In order for the Lwazi project to make an impact on the lives of South Africans, a service domain (e.g. Health, Education or Labour) and a specific application (e.g. an Automated Health Hotline or Bus Schedules) had to be selected based on an extensive survey of the information needs of the target audience. Once the application had been selected, the design had to take into account important human factors, such as the language and culture of the target group.
- Scientific and technical outputs development
The Lwazi information system showcases outstanding scientific and technical innovations, especially the creation of robust speech recognition (ASR) and text-to-speech (TTS) systems for all 11 official languages of South Africa. The integration of these language technologies into a telephony platform allows individuals to interact with the system by voice over a standard telephone line.
- Electronic linguistic resource collection
The Lwazi project, and future speech-based applications in South Africa, depends on the creation of extensive electronic linguistic resources both to generate and recognize speech. For each South African language, a pronunciation dictionary, an ASR corpus, and a TTS corpus is generated. An electronic repository enables the sharing of these valuable resources with the larger HLT research and development community.
|
