Inclusivity and Accessibility Focused Text Summarization
The proposed work presents a text summarization tool that is accessible and aimed at increasing the inclusivity and availability of information for users with diverse needs, especially people with disabilities or language barriers. By integrating text summarization, text-to-speech (TTS), multilanguage translation, and accessible graphical user interface (GUI) shortcuts, the tool shall provide an adaptable and userfriendly way to access complex information. The process uses a Hierarchical Attention Network (HAN) and a multi-task model known as Text-To-Text Transfer (T5) which would therefore ensure coherence in the contextual analysis as well as accuracy. The feature that enables the conversion of text into speech is quite supportive of multiple languages, thereby offering a far more natural audio replacement that would be of great benefit to the visually impaired, challenged readers, and auditory learners. In addition, the translation module allows users to view summaries in their preferred language, promoting linguistic inclusivity. Designed with accessible navigation shortcuts and in accordance with Web Content Accessibility Guidelines (WCAG) standards, the tool effectively accommodates users with motor impairments and guarantees compatibility with assistive technologies, such as screen readers. Significant challenges, including the necessity of balancing brevity with completeness and preserving contextual accuracy across various languages, are confronted through comprehensive testing. This project exemplifies a model for inclusive digital tools, promoting accessibility and nurturing an equitable information-sharing experience for all users.
Authors
- Renuka Devi S
- Nityanand Choudhary (ORCID: https://orcid.org/0000-0001-6099-9491)
- J. Harshini
- Nils C. Gerhardt (ORCID: https://orcid.org/0009-0002-5538-231X)
Institutions
- Vellore Institute of Technology University (IN)
Publication Details
- Published
- 2025-04-17
- DOI
- https://doi.org/10.1109/icdsbs63635.2025.11031904
- Primary Topic
- Natural Language Processing Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00