weekend project that uses the latest Chrome Summarizer API, but also YOLO to process PDF pages and find images, tables, and math formulas.
Readability (Mozilla) is used when you open regular web pages, but PDF.JS (also Mozilla!!) is used to render each page, run YOLO region detection and then generate a clean HTML. Math formulas are rendered as Latex, and tables are rendered into HTML by detecting their structure using Texo/FormulaNet.
Finally, you can read out loud the readable articles/PDFs using another local TTS (I prefer PocketTTS personally, but wanted to try the newly released and tiny Inflect TTS as well, so it's an option).
Let me know if you find this useful.
I should probably publish them to the Chrome and Firefox stores.
ldenoue•1h ago
Readability (Mozilla) is used when you open regular web pages, but PDF.JS (also Mozilla!!) is used to render each page, run YOLO region detection and then generate a clean HTML. Math formulas are rendered as Latex, and tables are rendered into HTML by detecting their structure using Texo/FormulaNet.
Finally, you can read out loud the readable articles/PDFs using another local TTS (I prefer PocketTTS personally, but wanted to try the newly released and tiny Inflect TTS as well, so it's an option).
Let me know if you find this useful.
I should probably publish them to the Chrome and Firefox stores.
ldenoue•1h ago