Touchscreen Keyboard Corrective Word Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing touchscreen keyboards for electronic devices are inefficient for character input, particularly for long messages, as users must divert attention from the keyboard to view suggested words, leading to eye strain and processing inefficiencies, and do not adequately accommodate different user typing patterns.
Innovation Solution
A method that predicts and displays sets of characters on the virtual keyboard without requiring users to divert attention, allowing rapid and precise typists to input characters efficiently by generating and ranking predicted sets of characters based on input and context, and displaying them at appropriate locations on the keyboard.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If suggested words are displayed separately from the keyboard, then word prediction functionality is provided, but users must divert attention from the keyboard causing eye strain and processing inefficiencies
Solution Approach 1:
The patent merges the word suggestion display with the keyboard interface by displaying suggested words at specific key locations on the keyboard itself. This integration allows users to view predictions without diverting their gaze from the keyboard, eliminating the need to shift attention between separate display areas and maintaining continuous focus on the input area.
Solution Approach 2:
The patent utilizes the vertical space above keyboard keys to display suggested words, adding a new dimensional layer to the keyboard interface. This approach transforms the traditional horizontal layout into a multi-layered display where predictions appear in the space above relevant keys, enabling information presentation without expanding the horizontal footprint or requiring separate display areas.
2Extent of automation
If users view suggested words displayed elsewhere on the display, then word prediction is available, but processing cycles are lost and display power is wasted
Solution Approach 1:
By combining the word suggestion display with the existing keyboard display area, the patent eliminates the need for separate display regions. This merging reduces the overall display active area, decreases processing overhead for managing multiple display elements, and minimizes energy consumption by utilizing already-active keyboard display regions rather than activating additional display zones.
3Area of stationary object
If the touchscreen area is small, then the device is portable, but character input becomes cumbersome particularly for long messages
Solution Approach 1:
The patent implements preliminary action by displaying predicted words before the user completes typing them. The system anticipates the user's intended input and presents suggestions in advance at relevant keyboard key locations, allowing users to confirm or correct predictions without completing the full character sequence. This reduces the number of keystrokes required and accelerates input speed for long messages.
Solution Approach 2:
The system provides self-service by automatically generating and displaying word predictions based on the user's partial input. The predictive text system continuously analyzes the user's typing and autonomously presents suggested completions, reducing the cognitive and physical effort required from the user and enabling faster input without requiring manual intervention or attention diversion.
Data Source
AI summary
The present disclosure provides a touchscreen keyboard with corrective word prediction. A method for correcting text input on an electronic device is described. The method comprises: displaying a virtual keyboard on a touchscreen, the virtual keyboard including a plurality of keys; receiving input from the virtual keyboard; generating one or more predicted sets of characters in accordance with the received input; and displaying a predicted set of characters at a designated location when the received input does not match one of the predicted sets of characters.