If you haven’t personally experienced the humiliation that is so often associated with iPhone’s autocorrect feature, you’ve probably heard from others that have. It’s an incredibly common occurrence that has sparked dozens of humor sites like autocorrectfail.org and others. Conversations like this:
Are constantly being “fixed,” most often before the person sending the message can catch it. Now, an Apple patent application “Combining timing and geometry information for typing correction” hopes to put a stop to the frequently mistaken autocorrect.
The patent details describe a technology that records where your fingers touch the screen as well as how fast and build a profile based on your habits. Each person’s unique typing habits are then put into a formula, allowing the program to discern whether you meant to type “we’ll” or “well” and other more extreme word pairings.
It wouldn’t completely eliminate the autocorrect feature, but it would ideally increase its efficiency. The application itself gives the example, “if the string ‘theere’ is entered and the time between the keystrokes ‘ee’ is less than an associated baseline by more than a threshold amount, the replacement candidate ‘there’ may be assigned a higher score.”
In conjunction with this patent is another which is able to use “parts-of-speech tagging” in order to determine the intended word(s). This means that not only would your phone be documenting your typing patterns, but it would also have the ability to determine the context around what is being typed.
While it’s not in the making yet, it’s probably safe to say that this application would be greatly appreciated by many and would keep iPhone users from experiencing these frequent texting “fails.”
Have an autocorrect story you'd like to share? We'd love to hear it!