Acoustic Stroke Input for Visually Impaired Accessibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for generating characters on electronic devices are limited in flexibility and may not be suitable for all environments, particularly for users who are visually impaired, as they require interaction with physical or virtual keyboards.
Innovation Solution
A method and electronic device that uses a conventional microphone in combination with text disambiguation software to receive input and output characters based on sound signatures of strokes written on a surface, allowing users to input characters without interacting with a keyboard, using a processor to detect and process sounds generated by strokes made with a finger, pen, or stylus.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional methods (physical keyboard, virtual keyboard, handwriting recognition, OCR, speech recognition) are used for character input, then character generation is effective, but flexibility and accessibility are limited especially for visually impaired users
Solution Approach 1:
The patent replaces mechanical keyboard interaction with acoustic field-based input. A microphone captures sound waves generated by writing instruments (pen, finger, stylus) moving across a surface, converting mechanical writing actions into electrical audio signals for processing. This eliminates the need for physical or virtual keyboard interaction while maintaining character input functionality.
Solution Approach 2:
The patent introduces sound waves as an intermediary between the user's writing action and the electronic device's character recognition system. The microphone acts as a mediator that captures the acoustic signature of writing strokes and transmits this information to the processor, which analyzes the sound patterns to identify written characters without direct keyboard interaction.
2Adaptability or versatility
If sound-based stroke recognition is implemented, then flexibility and accessibility are improved, but measurement precision of stroke identification must be maintained
Solution Approach 1:
The patent replaces visual-based handwriting recognition with acoustic-based stroke identification. Instead of capturing visual images of written characters and processing them through image recognition algorithms, the system captures sound waves generated during writing and analyzes acoustic patterns to identify strokes and characters, improving accessibility while maintaining identification accuracy through audio signal processing.
Solution Approach 2:
The patent changes the detection parameter from visual characteristics to acoustic characteristics. By analyzing sound wave properties such as frequency, amplitude, and temporal patterns generated during writing, the system can identify strokes and characters with high precision. This parameter transformation enables visually impaired users to write without visual feedback while the system accurately recognizes the intended characters through acoustic analysis.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables users to input and output characters without the need for a keyboard, enhancing flexibility and accessibility, especially for visually impaired individuals by converting sound signatures into recognizable characters.
Implementation Method 1
A method and electronic device that uses a conventional microphone in combination with text disambiguation software to receive input and output characters based on sound signatures of strokes
Data Source
AI summary
A method and electronic device for receiving input and outputting characters based on sound stroke patterns are described. In accordance with one embodiment, there is provided a method for receiving input on an electronic device, comprising: detecting sounds generated by characters written on a writing surface; identifying strokes defining the characters from the detected sounds in accordance with a number of predefined sound signatures, wherein each character written on the writing surface is defined by a stroke sequence comprising at least one stroke; determining at least one character which matches the at least one stroke in each stroke sequence; and outputting a string comprising the at least one character which matches the at least one stroke in each stroke sequence.


