Obook turns PDFs, ebooks and classics into natural-sounding audiobooks. Unlike basic text-to-speech readers, Obook first uses AI to clean the text (skipping headers, page numbers and footnotes) and splits it into real chapters, so you hear the book, not the page layout. Choose from 32 narrators across 20+ languages and follow along with word-by-word highlighting synced to the audio. Built and run on our own GPU voice pipeline by a solo founder.
Hey Product Hunt! 👋 I'm Emin, the solo founder behind Obook.
I built it because most text-to-speech apps just read whatever is on the page: page numbers, running headers, footnotes, broken line breaks. It sounds robotic and you lose the thread fast.
Obook works differently. When you upload a book, an AI pass cleans the text and organizes it into proper chapters before any audio is generated. Then it's narrated by one of 32 voices in 20+ languages, running on a GPU pipeline I built and host myself. While you listen, each word is highlighted in sync, so you can switch between reading and listening anytime.
It's available on iOS and Android. I'd love your honest feedback: which voices you like, what books break it, and what you'd want next. I'll be here all day answering questions! 🎧📚