Individually Customized Keyboard ― Highly Accurate Typing on Touchscreens
Typing on touchscreens lacks tactile feedback and therefore, is more error-prone and slower compared to regular hardware keyboards. This draw-back is currently compensated by applying algorithmic approaches like auto-completion, error correction and predictions based on statistical language models. Furthermore, key positions and key sizes are commonly adapted during runtime.
We extend and complement these approaches by taking account of the user’s fingerprints. The user’s individual typing behavior is analyzed utilizing the complete touch sensor data for each keystroke. Then, for each typist an individual keyboard model with individual key sizes is automatically generated. Our algorithm offers high text entry accuracy –- even for unknown words.
Further Information: PDF
Eberhard Karls Universität Tübingen
Phone: +49 (7071) 29-72639
Contact
Dr. Rolf Hecker
