UOB has signed an MoU with A*STAR to explore using MERaLiON in its relationship managers’ customer calls.
The bank will assess a MERaLiON-powered tool that interprets conversations across languages and checks afterwards whether RMs communicated clearly and met fair dealing standards.
According to UOB, it is the first bank to explore this solution, which A*STAR developed.
The first phase will run under Project AURA, the bank’s programme to deploy advanced AI at scale, where relationship manager (RM) enablement is one of its priority domains.
The MoU with the ASTAR Institute of Advanced Intelligence and Computing (ASTAR IAIC) will examine how MERaLiON’s capabilities could be enhanced to support UOB’s efforts to deploy the tool safely and at scale.
UOB is looking at its application in Singapore and across its five ASEAN markets.
How the MERaLiON solution works
The tool, formally named the Compliance Intelligence Engine and Evidence-Grounded Post-Call Compliance Review System, interprets customer questions and concerns across different languages, accents and communication styles.
During a call, it can prompt RMs on key product information and risk disclosures to complement their professional judgement.
After the call, it can conduct an evidence-grounded review to assess whether the RM communicated information clearly and met fair dealing standards, providing an additional layer of governance for the bank.
UOB said the system could improve productivity by reducing manual translation and automating parts of the post-call review process.
The insights it generates could also support more targeted coaching for RMs, who remain accountable for the conversations and advice they provide.

“Banking is built on trust, and every customer conversation matters. We see potential for this solution to help our RMs better understand customers, navigate language and cultural nuances, and provide clearer, more relevant guidance,”
said Leong Yung Chee, Group Chief Financial Officer, UOB.
Built for code-switching across ASEAN
UOB serves more than eight million customers across ASEAN, where customer conversations may move between languages, include local expressions or switch languages within the same exchange.
MERaLiON supports multilingual processing and code-switching across 11 ASEAN languages, including Malay, Bahasa Indonesia, Thai and Vietnamese, alongside English, Mandarin, traditional dialects such as Cantonese and Singlish.
The model has achieved the lowest average word error rate among open-source models across languages.
A*STAR collaboration builds on wider financial sector efforts
The MoU builds on a broader collaboration between A*STAR and the Future of Finance Institute, established by the Monetary Authority of Singapore (MAS), to develop and test AI technologies against real-world needs in the financial sector.

“Singapore’s multilingual AI research can create greater impact when it is tested against complex, real-world needs. Through this collaboration with UOB, we are exploring how MERaLiON can be applied in a demanding banking environment where accuracy, explainability and cultural context are critical,”
said Professor Lim Keng Hui, Assistant Chief Executive, Science and Engineering Research Council, A*STAR.
UOB said it will continue to assess how MERaLiON’s capabilities can be applied safely and responsibly across its regional franchise, while keeping its people at the centre of customer engagement.
Featured image credit: Edited by Fintech News Singapore, based on image by swampfire via Magnific


