Intelligent Content Parsing for Synthetic Speech and Braille

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing screen readers convert web content into speech and braille in a top-to-bottom manner, disregarding user relevance, limiting user control over content reproduction.

Innovation Solution

An intelligent content parsing and synthetic speech and braille production system that identifies and sorts tag sequences based on user preferences, generating summaries and outputs relative to expected reading times for audio and braille reproduction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If top-to-bottom content reproduction method is used, then content is reproduced in predetermined flow, but user control over content reproduction is limited and relevant content cannot be accessed efficiently

Engineering Contradiction:

Improveuser control over content reproductionVSAvoidcontent parsing and sorting system

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The patent segments web page content into distinct tag sequences (e.g., headers, paragraphs, lists, images) and processes them independently. Each tag sequence is identified, analyzed for relevance, and sorted separately, allowing users to access specific content segments without processing the entire page sequentially. This segmentation enables selective reproduction of only the most relevant content sections.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary parsing and relevance analysis of all tag sequences before actual content reproduction. By pre-identifying and sorting tag sequences based on user profiles and content importance in advance, the system prepares content for optimized reproduction, allowing users to skip directly to relevant sections without waiting for sequential processing.

Inventive Principle:

Principle #10Preliminary action

2Loss of time

If all content is reproduced in predetermined flow, then complete content coverage is achieved, but time to access relevant content increases

Engineering Contradiction:

Improvetime to access relevant contentVSAvoidcontent completeness

Core Design Contradiction:

Loss of timeVSLoss of information

Solution Approach 1:

The patent applies different processing qualities to different tag sequences based on their relevance and content type. High-priority tag sequences (e.g., main headings, key paragraphs) are processed and reproduced first with full detail, while lower-priority sequences are processed subsequently or summarized. This local quality differentiation ensures critical information is delivered immediately while maintaining option for complete content access.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system performs partial content reproduction by selecting and reproducing only the most relevant tag sequences based on user profiles and content analysis, rather than reproducing all content. This partial action approach significantly reduces access time for relevant information while providing options for users to access additional content if needed, balancing speed with completeness.

Inventive Principle:

Principle #16Partial or excessive action

3Adaptability or versatility

If user-specific content sorting is implemented, then personalized content delivery is achieved, but processing complexity increases

Engineering Contradiction:

Improveuser-specific content sortingVSAvoidparsing and sorting system

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system automatically analyzes user profiles, preferences, and behavior patterns to self-determine the relevance and sorting order of tag sequences without requiring manual user configuration. The system serves itself by making intelligent decisions about content prioritization based on accumulated user data, reducing the apparent complexity for users while maintaining personalized content delivery.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system incorporates feedback mechanisms where user interactions with reproduced content (e.g., which sections are accessed, time spent on specific content) are analyzed and used to refine future content sorting and prioritization. This feedback loop enables the system to adapt to user preferences over time, improving personalization while the feedback processing is handled automatically by the system.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS10600337B2Intelligent content parsing with synthetic speech and tangible braille production

Publication Date: 2020.03.24 BANK OF AMERICA CORP

AI summary

Aspects of the disclosure relate to parsing page content to determine tag sequences and synthesizing content associated with the determined tag sequences to produce audio and/or braille output relative to user preferences and input. In a first embodiment, a user computing device may receive a page document corresponding to a uniform resource locator (URL) of the third party computing platform, identify one or more tag sequences of the page document, calculate an expected reading time for each of the one or more tag sequences, generate a summary associated with each of the one or more tag sequences of the page document, and produce an output of the summary. In a second embodiment, a server infrastructure may activate an interface with the user computing device and may perform the aforementioned processes in order to increase processing efficiency and decrease computing load at the user computing device.