![]() The result? SwiftKey users type one third fewer keystrokes than they do typing each letter individually. To watch an experienced Swiftkey user type out a text – fingers flying across the screen, sliding through sequences of letters and occasionally tapping a predicted word or phrase – can be a dizzying experience. Using the software’s Flow feature, you don’t even need to pick your finger up off of the screen. For instance, if you tend to hit the key just to the left of the one you want to, the app will learn that about you. “If you think about the problem of people trying to type on phones, it’s actually more of a problem about language than it is a problem about keyboards,” Medlock told FORBES.īuilt by analyzing passages from publicly available web text, the SwiftKey algorithm predicts your next words, corrects spelling, traces your finger placement on the screen, detects what language you are typing and snaps to it (60 languages are available), and learns a user’s individual texting quirks. Word prediction on phones and tablets is nothing new, but SwiftKey’s predictive typing technology is a cut above the rest, powered by machine learning principals studied by Medlock, 34, a Ph.D in natural language processing from Cambridge, and developed by a SwiftKey staff that boasts 17 other Ph.Ds in language, machine learning and big data.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |