Text-to-Face is a Google Glass hack which utilizes the Tesseract OCR and was inspired by the Spritz Speedreader. It works by taking photos of text with the Google Glass camera and sending it over network to a server. The server then runs the Google Tesseract OCR over the image to perform computer vision and extract a plaintext string, which it then sends back to the Google Glass. Using an algorithm following the methods of Spritz, the Glass then splits the string and presents the user with one word at a time, following their desired rate.
We were inspired by the sheer effectiveness of the Spritz system, which can double or triple online reading speeds. The portability and ease of use of the Google Glass camera seemed like a natural partner for bringing such a system into the real world.
Our target users are students and reading enthusiasts, who have a lot to read in very little time (or, in college students' cases, a ton to read in even less time). After all, the software works best with books.
We're very proud of the whole product; we think it's pretty novel, even if it is just translating visual text to visual text at the end of the day. If we had to pick a single feature, we're probably most proud of the crisp look and feel when the speedreading actually kicks in.