The North-West University has established a resource management agency (RMA) to manage reusable digital text and speech resources for all the country’s official languages.

The language resources are used in automated telephone information systems, global positioning systems, spelling and grammar checkers and machine translation systems applied in various South African language sphere, besides other uses.

“These language resources are crucial for research and development in the domain of human language technologies (HLT) for application in local languages. These types of applications are increasingly being used on mobile devices, giving South Africans unprecedented access to information and services in local languages,” the university said in a statement.

Broad categories of data relevant to the RMA include text, speech, language-related videos, multimodal resources, such as sign language, as well as pathological and forensic language data.

“South Africa and India are the only countries in the world that have a specific strategy for developing language technologies that are intended to facilitate communication among language groups in their respective societies. The founding of the RMA is the realisation of a long-awaited ideal of academics, language practitioners and developers to empower local languages for use in modern communication systems,” says North-West University professor Justus Roux.

The Department of Arts and Culture’s National Centre for Human Language Technologies is funding this initiative, which is in line with similar global initiatives.

The North-West University’s Centre for Text Technology (CTexT), Potchefstroom campus, has been appointed to establish the RMA and to distribute reusable text and speech resources from this central point.

“The Department of Arts and Culture has invested significantly in HLT over the last decade to stimulate activity and promote advances that may benefit our marginalised languages,” says Department of Arts and Culture Director of Human Language Technologies Ulrike Janke.

“Were it not for our implementation agencies, however, the return on investment would not have been so impressive. They are passionate about HLT and its potential and, therefore, continue to do their best to help us succeed in our strategy. We are thankful for productive partnerships such as this one with the North-West University,” she concludes.

Story highlights:

* The languages resources agency was established to enable the use of official languages in modern communication systems.

* Language data relevant to the agency include text, speech, language-related videos and multimodal resources, such as sign language, as well as pathological and forensic language data.