Gesture Keyboard Out-of-Vocabulary String Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gesture-based keyboards on computing devices struggle to accurately identify and output out-of-vocabulary character strings, often autocorrecting or discarding them, which increases user effort and errors when entering unique words, slang, or abbreviations.
Innovation Solution
A method that determines a character string selection based on both spatial and language model scores, associating out-of-vocabulary strings with a language model score and combining it with a spatial model score to indicate the probability of the gesture input, allowing for the output of character strings not included in the device's vocabulary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the computing device uses word prediction and auto-correction techniques based on a predetermined vocabulary, then text entry speed is improved and spelling mistakes are minimized, but out-of-vocabulary character strings cannot be correctly determined
Solution Approach 1:
The system changes the parameter of language model scoring by introducing out-of-vocabulary language model scores for character strings not in the predetermined vocabulary. This allows the system to evaluate and select OOV character strings based on gesture input patterns while maintaining the ability to handle both dictionary and non-dictionary words effectively.
2Reliability
If the computing device relies on dictionary-based word prediction techniques, then spelling mistakes are minimized, but additional user effort is required to enter non-dictionary character strings
Solution Approach 1:
The system makes the gesture keyboard universally applicable to both dictionary and out-of-vocabulary character strings. By implementing out-of-vocabulary language model scoring, the same gesture-based input mechanism works effectively for all character strings regardless of whether they exist in the predetermined vocabulary, eliminating the need for different input methods.
3Productivity
If the computing device uses traditional gesture keyboard techniques with predetermined vocabulary, then input efficiency is improved, but errors increase when entering unique words, slang, or abbreviations
Solution Approach 1:
The system implements feedback mechanisms by calculating out-of-vocabulary language model scores based on gesture input patterns and comparing them against spatial model scores. This feedback loop allows the system to continuously evaluate and select the most probable character string, whether in-vocabulary or out-of-vocabulary, thereby maintaining high accuracy while preserving input efficiency.
Data Source
AI summary
In one example, a computing device may include at least one processor that outputs a graphical keyboard. The at least one processor may determine a selection of one or more keys; and determine a character string based on the one or more keys in the selection, wherein the character string is not included in a vocabulary used by the computing device to identify character strings. The at least one processor may, in response to determining that the character string is not included in the vocabulary used by the computing device to identify character strings, associate the character string with an out-of-vocabulary language model score; and determine, based at least in part on a spatial model score associated with the gesture and the out-of-vocabulary language model score, a combined score that indicates a probability that the character string corresponds to the gesture; and output for display the character string.