Screen Reader Text Filtering for Power-Saving Speech Output
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Screen readers consume significant processing resources and power when dealing with large amounts of text, including words and alternative text associated with images and multimedia content.
Innovation Solution
Implementing a system that identifies and skips certain subsets of text, such as words corresponding to specific parts of speech and those with inadequate font sizes, using machine learning models to determine relevance, thereby reducing the amount of text processed by the text-to-speech algorithm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the text-to-speech algorithm processes all text content including alternative text and multimedia content, then complete information is provided to users, but processing resources and power consumption increase significantly
Solution Approach 1:
The patent segments text content into different categories (navigational elements, descriptive text, alternative text, multimedia content) and applies different processing strategies to each segment. The screen reader identifies and processes only relevant segments based on user needs, rather than processing all text uniformly, thereby reducing power consumption while maintaining information completeness.
Solution Approach 2:
The patent applies local quality by treating different portions of text content with different levels of processing intensity. Important content such as navigational elements and key descriptive text receives full processing, while less critical content like alternative text for decorative images or redundant multimedia descriptions is skipped or processed at lower fidelity, optimizing the balance between information completeness and power consumption.
2Reliability
If the screen reader processes all words in the set, then complete text coverage is achieved, but processing time and resource consumption increase
Solution Approach 1:
The patent performs preliminary analysis of the text content before full processing, identifying navigational elements, headings, and structurally important text in advance. This preliminary action allows the screen reader to prioritize processing of critical content while skipping or deferring less important content, thereby improving processing efficiency without sacrificing essential text coverage.
Solution Approach 2:
The patent extracts and processes only the essential subset of text content that provides meaningful information to users. By taking out and processing only navigational elements, key descriptive text, and structurally important content while excluding redundant alternative text and multimedia content, the system achieves efficient processing with adequate text coverage.
3Loss of information
If all text content is converted to speech, then users receive complete information, but power consumption and processing load increase
Solution Approach 1:
The patent implements dynamic processing where the screen reader adapts its text processing behavior based on user context, device state, and content importance. Users can dynamically adjust processing depth, and the system dynamically selects which text content to process based on real-time conditions, optimizing the balance between information completeness and power consumption for different usage scenarios.
Data Source
AI summary
In some implementations, a user device may receive a set of words, for conversion to speech, including a plurality of subsets of the set of words, where each subset of the plurality of subsets is associated with one or more corresponding formatting properties. The user device may identify, within the set of words, a first subset of the plurality of subsets as relevant based on the one or more corresponding formatting properties associated with the first subset. Additionally, the user device may identify, within the set of words, a second subset of the plurality of subsets as not relevant based on the one or more corresponding formatting properties associated with the second subset. Accordingly, the user device may input the first subset to a text-to-speech algorithm.