Adaptive Font Rendering for Vocal Content Transcripts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Deaf or hard-of-hearing individuals are unable to fully experience the oral characteristics of vocal content, such as prosodic features, linguistic characteristics, and paralinguistic features, when relying on plaintext transcripts.
Innovation Solution
A computer-implemented method that adapts the font of textual information to convey oral characteristics of vocal content by determining feature parameters associated with vocal features using a trained model and applying these parameters to select or generate an adaptive font.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If plaintext transcripts are used to convey vocal content, then the informational literal content is conveyed, but the oral characteristics (prosodic, linguistic, and paralinguistic features) are lost
Solution Approach 1:
The patent applies parameter changes by mapping vocal feature parameters (prosodic, linguistic, paralinguistic) to font parameters (size, weight, style, color). The trained model extracts parameters from audio data and transforms them into corresponding font formatting parameters, enabling the transcript to visually represent oral characteristics without fundamentally changing the transcript structure
Solution Approach 2:
The patent introduces an intermediary system consisting of a trained model and font adaptation mechanism that bridges the gap between audio data and text transcript. This intermediary extracts oral characteristics from audio and translates them into visual font representations, allowing the transcript to convey both literal and oral information without directly modifying the audio or text themselves
2Loss of information
If adaptive fonts are used to convey oral characteristics, then full access to vocal content is enabled, but the complexity of the system increases
Solution Approach 1:
The system uses parameter changes by mapping vocal feature parameters to font parameters through a trained model. This allows the representation of oral characteristics using existing font parameter spaces, avoiding the need for entirely new complex systems while still achieving comprehensive conveyance of vocal content
Solution Approach 2:
The adaptive font system serves multiple functions simultaneously: it conveys literal text information, prosodic features, linguistic characteristics, and paralinguistic features all through a single unified font formatting mechanism. This multi-functionality reduces the need for separate systems for each type of information representation
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The disclosure relates to methods, devices, systems and media for providing and using an adaptive font to convey oral characteristics of utterances comprised in audio data in a transcript of the utterances. For instance, a method may comprise obtaining audio data, obtaining textual information transcribing an utterance comprised in the audio data, determining for the utterance, using a trained model, a plurality of feature parameters associated with a plurality of vocal features, determining an adaptive font based on the plurality of feature parameters, and formatting at least a part of the textual information using the adaptive font for display to a user. The plurality of features parameters may be peculiar to the utterance. Various techniques are disclosed to determine the adaptive font based on the plurality of feature parameters.