Google Maps is improving its transliteration system to display better results when searched with native Indian scripts and not English. Google says that as India's internet population is expected to grow from 75 percent to 90 percent in the next five years, it wants its app to be as accessible as possible even when users are searching in their preferred language. Transliteration is different from translation as it means writing the same words in another script without losing its relevance. Google says that the new transliteration system will help many users who do not speak English, adding that most Indian places of interest (POIs) on Google Maps are not available in regional language.
On its blog, Google announced the addition of automatic transliteration for 10 Indian language that includes Bangla, Gujarati, Hindi, Kannada, Malayalam, Marathi, Odia, Punjabi, Tamil, and Telugu. To deal with the tricky acronyms, the company has developed a specialised transliteration module that generates additional candidate transliterations for unique cases. Here's how the company explains the problems faced by Google Maps users in India while looking for places in their regional language:
"Consider a user in Ahmedabad, Gujarat, who is looking for a nearby hospital, KD Hospital. They issue the search query, કેડી હોસ્પિટલ, in the native script of Gujarati, the sixth most widely spoken language in India. Here, કેડી (kay-dee) is sounding out of the acronym KD, and હોસ્પિટલ is 'hospital.' In this search, Google Maps knows to look for hospitals, but it doesn't understand that કેડી is KD, hence it finds another hospital, CIMS. As a consequence of the relative sparsity of names available in the Gujarati script for places of interest (POIs) in India, instead of their desired result, the user is shown a result that is further away."
Google explains that the new transliteration system combines scores for the possible transliterations in "weighted mixtures" that are tuned specifically for POI name accuracy using small targeted development sets for such names. Under the improved mode, Google Maps now displays 19 times better Bengali, Tamil and Marathi results, while Hindi results have also improved 3x. As with any machine-learned system, the resulting automatic transliterations may contain a few errors but the large increase in coverage in these widely spoken languages marks a substantial expansion of the accessibility of information within Google Maps in India. Google promises to improve its system in future.