Jul 12, 2021The Future of Neural Machine TranslationImprovements to Neural Machine Translation (NMT) engines fall into one of two buckets. The first bucket has to do with accuracy. These problems are largely met with engineering solutions that offer incremental improvements. This applies to high-resource language pairs (e.g., English to Spanish), but especially to low-resource language pairs (e.g…Translation2 min read
Jul 12, 2021Neural Machine Translation in Messaging Applications TodayGiven the complexity of Neural Machine Translation (NMT), it will come as no surprise that every major messaging app has opted to “buy” rather than “build” their translation capabilities. …Translation2 min read
Jul 12, 2021Landscape Analysis: Neural Machine TranslationThe Big 3, when it comes to neural machine translation (NMT), are Google, Microsoft, and Amazon. Among this group, Google is the most dominant in terms of supporting 109 languages compared to Microsoft’s 73, and Amazon’s 55. …Translation3 min read
Jul 1, 2021Three Challenges with Neural Machine TranslationChallenge 1: Models can’t see the forest for the trees While Neural Machine Translation (NMT) is leaps and bounds ahead of where the industry was a couple of decades ago, it still has its limitations. Over time we’ve seen translation models zooming out from the individual word level to the sentence level, which is where they are stuck today. …Translation2 min read
Jul 1, 2021A brief history of Machine TranslationThe origins of machine translation date back to the early days of The Cold War. Inspired by the success of code-breaking efforts in WWII, American scientists attempted to “unscramble” Russian text. …Translation2 min read
