Eye Tracking AAC System with Adaptive Text Suggestions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Individuals relying on eye gaze for communication face significant delays due to the slow process of selecting icons and formulating speech using existing AAC systems, which are often unsuitable for personal speech patterns and require time-consuming navigation and editing.
Innovation Solution
An AAC system using an eye tracking device that provides adaptive suggested text based on message content, allowing users to select and modify text through gaze inputs, with features like navigation buttons for efficient cursor movement and visual indications for selection feedback.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If eye gaze technology is used to select icons for communication, then accessibility for non-verbal users is improved, but communication speed deteriorates significantly
Solution Approach 1:
The system performs preliminary actions by providing predictive text suggestions and automatically completing phrases based on historical data and context analysis. This allows users to select pre-formed phrases rather than constructing each message character-by-character, dramatically reducing the number of gaze selections needed while maintaining accessibility.
Solution Approach 2:
The system copies and reuses historical speech patterns, phrases, and communication styles specific to each user. By analyzing past communications and creating personalized phrase banks, the system enables users to communicate more efficiently by selecting from pre-compiled phrases that reflect their personal speech patterns, rather than manually constructing each message.
2Productivity
If multiple phrase suggestions are provided to users, then communication efficiency is improved, but the time spent checking and selecting suggestions increases
Solution Approach 1:
The system applies local quality by providing context-specific suggestions at the appropriate location in the message composition area. Instead of displaying multiple generic suggestions that require scanning, the system provides targeted suggestions relevant to the current context directly where the user is typing, reducing the cognitive load and time required to evaluate options.
Solution Approach 2:
The system implements feedback mechanisms that adapt to user preferences and communication patterns in real-time. By analyzing which suggestions are accepted or rejected and adjusting the suggestion algorithm accordingly, the system learns to provide more accurate and relevant suggestions over time, reducing the time users spend evaluating options while maintaining high communication efficiency.
3Ease of operation
If pre-set phrases are provided in the system, then initial usability is improved, but personalization and accuracy of speech representation deteriorate
Solution Approach 1:
The system transitions from static pre-set phrases to dynamic, adaptive phrase generation. It continuously learns from each user's communication patterns, speech style, and phrase preferences, automatically updating and personalizing the phrase bank over time. This allows the system to maintain ease of use while progressively improving personalization accuracy as it adapts to each user's unique communication style.
Solution Approach 2:
The system performs self-service by automatically analyzing and learning from user communications without requiring manual programming or configuration. It autonomously builds personalized phrase banks by monitoring communication patterns, extracting frequently used phrases and speech patterns, and continuously refining its suggestions, thereby achieving personalization while maintaining ease of operation.
4Ease of operation
If cursor navigation through speech text is required for editing, then editing capability is improved, but time consumption increases significantly
Solution Approach 1:
The system extracts and highlights the specific portion of text that requires editing or modification based on contextual analysis. Instead of requiring users to manually navigate through entire sentences or paragraphs to locate editing points, the system automatically identifies and extracts the relevant text segment, presenting it prominently for quick selection and modification, thereby maintaining editing capability while dramatically reducing navigation time.
Data Source
AI summary
Provided are methods, systems, and computer program products for using an augmented or alternative communications (AAC) system including an eye tracking device. The method may include receiving at least one gaze input from a user and displaying at least one selected text portion in response to each gaze input. The at least one selected text portion may form a message that is displayed in a message window of the AAC system. A suggested text portion may be generated, the suggested text portion being adaptive to the content of the message and being displayed in the message window with the message. The method may further include receiving a confirmatory gaze input from the user, the confirmatory gaze input causing the suggested text portion to be added to the message.