$\varepsilon$ KÚ <MASK>: Integrating Yorùbá cultural greetings into machine translation

This paper investigates the performance of massively multilingual neural machine translation (NMT) systems in translating Yor\`ub\á greetings ($\varepsilon$ k\ú [MASK]), which are a big part of Yor\`ub\á language and culture, into English. To evaluate these models, we present IkiniYor\`ub\á, a Yor\`ub\á-English translation dataset containing some Yor\`ub\á greetings, and sample use cases. We analysed the performance of different multilingual NMT systems including Google and NLLB and show that these models struggle to accurately translate Yor\`ub\á greetings into English. In addition, we trained a Yor\`ub\á-English model by finetuning an existing NMT model on the training split of IkiniYor\`ub\á and this achieved better performance when compared to the pre-trained multilingual NMT models, although they were trained on a large volume of data.