Show HN: Readability for web and PDFs with TTS and local AI summary (github.com)

A new browser extension called Readability Reader has been launched, offering a private, local-first solution for reading articles and PDFs with AI-generated summaries and text-to-speech (TTS) capabilities. Available for Chrome and Firefox, this extension streamlines content consumption by providing a clean reading view, local extraction of article text using Mozilla Readability, and PDF conversion while maintaining text hierarchy through PDF.js. It also employs advanced layout analysis via a bundled DocLayNet model, which identifies text regions, images, tables, and formulas, converting them to LaTeX for display. This development is significant for the AI/ML community as it emphasizes privacy—processing everything locally without sending data to servers—and showcases the integration of various AI technologies. With the ability to generate concise on-device summaries and read aloud with customizable voices, Readability Reader enhances accessibility and focus for users. The extension also features caching for efficiency, ensuring that required AI models and voice files are stored for quick retrieval in future sessions. Overall, it represents a noteworthy step in the evolution of web and PDF reading tools, leveraging local AI capabilities to improve user experience while safeguarding privacy.