Minister for Culture, Lands and Local Government Owen Bonnici has announced an innovative project called SaqWI-QA/SaqWI which aims to strengthen the digital presence of Maltese by developing essential resources for natural language processing.
Supported by the Microsoft funding scheme: Lingua, SaqWI has been awarded $51,000, including $15,000 in Azure Credits. The scheme is dedicated to helping under-resourced European languages on AI platforms, with the Maltese Language Centre being the beneficiary. This was announced during a meeting between Minister Owen Bonnici and Microsoft General Manager for Malta, Greece and Cyprus Yana Andronopoulou.
Also present at the meeting were Microsoft National Technical Officer Katerina Sourouni, Microsoft Malta Technology Strategist Kyle Anastasi, Head of the Maltese Language Centre Dr Maris Camilleri, together with the University of Malta’s Department of Artificial Intelligence and the NLP Research Group led by Prof. Claudia Borg, PBS CEO Eng. Keith Chetcuti and the CIO of the Ministry for Culture, Arts and Local Government Christian Attard.
Minister Owen Bonnici welcomed this news with satisfaction and stated that the SaqWI project is an important step forward for the technological development of the Maltese language, while ensuring its presence and relevance for generations to come. He also emphasised that collaborations with public and private entities will continue to be strengthened so that Maltese can move forward in the digital world.
The project will use the extensive audiovisual archives of the Public Broadcasting Services (PBS) to create an open dataset that includes at least 5,000 sets of questions and answers and 100–150 hours of aligned transcript snippets. These resources, based on local broadcasts such as news, documentaries and cultural programmes, ensure that the dataset is culturally and historically relevant to the Maltese population.
Azure AI services will be used to transcribe the broadcasts, with human annotators refining the output for cultural and linguistic accuracy. SaqWI will provide tools for future educational and technological applications. The dataset will be released under open licenses, to encourage long-term sustainability and further research on Maltese as an underrepresented language.
It will also serve to benchmark and audit the quality of automated speech transcription (ASR), and improve transcription for underrepresented languages. In addition, the project will collaborate with TVM and Microsoft to improve digital ASR tools for Maltese, using the 6pm news as a pilot project. This will address challenges such as code-switching, spelling and intonation variation, and open the door to inclusive subtitling solutions for the deaf, the elderly, language learners and the general public.
