Learning languages and translation apps
accuracy of their predictions. 14 This means that, given a vast dataset, NMT can learn quickly and adapt to new contexts without human intervention. This results in high accuracy while being more cost effective than a human translator. Is this the whole story?
Machine translation currently works best in situations rooted in objective reality, that is, when confronted with clear mathematical or physical rules that govern their decision-making, such as writing manuals. 15 Natural language, although rule-based, is a subjective entity in which there is nuance, wit, wordplay, context, and by its very nature is always evolving. Paul Ricoeur argued that ‘ language could extend itself to its very limits forever discovering new resonances within itself ’. 16 Language relies on human wisdom to evolve. AI and human wisdom are essentially two different things, with the former acting as the result of the latter. If I find myself in a Pashtun home, I may turn to Google Translate to accurately decipher what my host has just asked me. Learning that he has asked me to dinner, I may be minded to say, Yes. But my human wisdom and cultural sensitivity teaches me that it would be offensive to my sword-wielding host to accept his offer of food at the first asking. I must engage in a back-and- forth of polite refusal. But Google Translate doesn’t come with this footnote. Language is an art more than it is a science, an evolving entity more than a rule-based yardstick. Therefore, it will always be the preserve of the unpredictable, evolving, wise species that is the human being. What we can do is to harness its strengths and ours to push the boundary of language. Most experts agree, 17 therefore, that the future of translation will combine NMT and human capabilities — a future where machines will bring scalable capacity to translation while humans will provide creativity, critical thinking, and nuanced interpretation. Translation apps have their very useful place in communicating with a greater range of people than we have ever done before. But the wide range of benefits to individual and society of learning language are too vast to outsource to Google, and humanity would be the poorer for it.
Additional bibliography
Habash, F. ‘ AI Translation Vs. Human Translation: Pros And Cons ’, at https://www.getblend.com/blog/ai- translation-vs-human-translation-pros-and-cons. Consulted: 19/7/2024 Mohamed, Y. et al. (2024) ‘The impact of artificial intelligence on language translation: a review’ , IEEE Access 12, https://ieeexplore.ieee.org/document/10438431. Consulted: 19/7/2024 Ordorica, S. ‘Why Technology Will Not Replace Professional Translators’, https://www.forbes.com/sites/forbesbusinesscouncil/2020/10/26/why-technology-will-not-replace- professional-translators/.Consulted: 19/7/2024 Patel, H. ‘ Risks and consequences of ai translation in 2023 ’, https://translatebyhumans.com/blog/risks-and- consequences-of-ai-translation-in-2023/. Consulted: 19/7/2024 https://www.betranslated.com/blog/ai-machine-learning-translation/ Consulted: 19/7/2024 14 https://blog.purestorage.com/purely-educational/deep-learning-vs-neural-networks/ Consulted: 19/7/2024. 15 Rechtman, J. AI Won’t Replace Human Translators Yet. Here Are 3 Reasons Why https://www.weforum.org/agenda/2018/10/3-reasons-why-ai-wont-replace-human-translators-yet/ Consulted: 19/7/2024. 16 Luo, X. (2018) ‘Artificial intelligence and the crisis of translation’, Asia Pacific Translation and Intercultural Studies 5.1: 1-2. https://www.tandfonline.com/doi/full/10.1080/23306343.2018.1456440. Consulted: 19/7/2024. 17 Chan, S. (2017) The Future of Translation Technology: Towards a World Without Babel . Oxford.
232
Made with FlippingBook flipbook maker