Also written by Devin Coldewey:
Microsoft hits a speech recognition milestone with a system just as good as human ears | Follow @2Elearning
x hide permanently

Google’s Transformer solves a tricky problem in machine translation


Machine learning has turned out to be a very useful tool for translation, but it has a few weak spots. The tendency of translation models to do their work word by word is one of those, and can lead to serious errors. Google details the nature of this problem, and their solution to it, in an interesting post on its Research blog.