AAC Text Prediction Interface for Low-Input Word Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AAC systems face challenges in efficiently enabling individuals with neuromuscular disorders to communicate effectively due to the complexity of virtual keyboards and the need for precise input, especially for languages with numerous keyboard variants, leading to increased effort and potential errors in word entry.
Innovation Solution
The implementation of an AAC system utilizing an Accessible Shell framework that includes an eye-tracking interface, n-gram dictionaries, and a novel text prediction mechanism using Large Language Models (LLMs) to facilitate efficient word selection and entry through a simplified user interface, reducing the number of input actions required and minimizing errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a virtual keyboard is used for text entry in AAC systems, then flexibility in creating text is improved, but the complexity of the interface and the number of input actions required increases
Solution Approach 1:
The virtual keyboard is segmented into multiple pages with different keyboard layouts (QWERTY, AZERTY, Dvorak, etc.). Users can navigate between pages to select the appropriate keyboard variant for their language or preference, reducing the complexity of any single page while maintaining overall versatility.
Solution Approach 2:
The keyboard interface is made dynamic through predictive text functionality. As users type or select words, the system dynamically predicts and suggests likely next words based on context, allowing users to confirm, correct, or navigate to alternative predictions. This reduces the number of manual input actions required while maintaining flexibility.
2Adaptability or versatility
If a virtual keyboard with many language variants is provided, then adaptability to different languages is improved, but the time required to select and enter words increases
Solution Approach 1:
The system performs preliminary action by pre-configuring multiple keyboard layouts (QWERTY, AZERTY, Dvorak, etc.) and having users select their preferred variant before text entry begins. This preliminary selection eliminates the need to switch between keyboard variants during the communication process, reducing time loss.
Solution Approach 2:
Predictive text provides feedback to users as they type, suggesting likely next words based on the context so far. Users can confirm predictions with a single action, correct errors, or navigate to alternative predictions. This feedback mechanism significantly reduces the time required for word entry compared to traditional virtual keyboards.
3Measurement precision
If precise input is required for accurate word entry, then communication accuracy is improved, but the difficulty of operation increases for users with limited motor skills
Solution Approach 1:
The predictive text system provides continuous feedback by suggesting likely next words as users input. This feedback mechanism helps users with limited motor skills achieve accurate word entry by reducing the precision requirements for each individual input action while maintaining overall communication accuracy through context-aware predictions.
Solution Approach 2:
The system performs self-service by automatically predicting and suggesting words based on the conversation context. This reduces the burden on users with limited motor skills, as the system handles the complex task of determining the next word, while users only need to confirm or make minor corrections.
4Adaptability or versatility
If multiple keyboard variants are supported, then language adaptability is improved, but the effort to select the correct keyboard increases
Solution Approach 1:
Users perform preliminary action by selecting their preferred keyboard variant before text entry begins. Once selected, the keyboard layout remains fixed for the duration of the communication session, eliminating the need to repeatedly select between variants. This preliminary selection significantly reduces the effort required during actual word entry.
Data Source
AI summary
Method, software, and apparatus for improved Augmentative and Alternative Communication (AAC) solutions. In one aspect, a user interface is provided with a set of suggestions comprising text, phrases, etc., and navigation buttons that enable users to select words and phrases to add to be written and/or spoken in a manner that reduces the number of user inputs. The suggestions are displayed in alphabetical order in rows with navigation buttons adjacent to the rows, with activation of a navigation button resulting in generation of updated suggestions having alphabetical ranges that are bounded by suggestions in associated rows. This approach may be combined with predictive text means to enable users to easily formulate text and/or speech content.