Algorithm learns how to revive lost languages

时间:2019-03-01 03:16:01166网络整理admin

By Douglas Heaven LIKE living things, languages evolve. Words mutate, sounds shift and new tongues arise from old. Charting this landscape is usually done through manual research. But now a computer has been taught to reconstruct lost languages using the sounds uttered by those who speak their modern successors. “The system reconstructs lost languages based on the sounds of phonetically similar modern tongues” Alexandre Bouchard-Côté at the University of British Columbia in Vancouver, Canada, and colleagues have developed a machine-learning algorithm that uses rules about how the sounds of words can vary to infer the most likely phonetic changes behind a language’s divergence. For example, in a recent change known as the Canadian Shift, many Canadians now say “aboot” instead of “about”. “It happens in all words with a similar sound,” says Bouchard-Côté. The team applied the technique to thousands of word pairings used across 637 Austronesian languages – the family that includes Fijian, Hawaiian and Tongan. The system was able to suggest how ancestor languages might have sounded and also identify which sounds were most likely to change. When the team compared the results with work done by human specialists, they found that over 85 per cent of suggestions were within a single character of the actual words (PNAS, 10.1073/pnas.1204678110). The technique could improve machine translation of phonetically similar languages. Endangered languages could also be preserved if they are phonetically related to more widely spoken tongues, says Bouchard-Côté. He is now working on an online version of the tool for linguists to use. This article appeared in print under the headline “Ancestor algorithm revives languages of the past” More on these topics: