Built an EPUB Text-to-Speech Reader using hashtag#Supertonic3 — a lightning-fast and much better than the earlier version, so the on-device TTS model that runs entirely via hashtag#ONNX Runtime. No hashtag#OpenAI, no hashtag#ElevenLabs, no per-character billing.
What it does: → Upload an hashtag#EPUB or paste any text → Converts it to sentences and generates audio on the fly → Click any sentence to jump to it → 7 voice styles, 31 languages, adjustable speed → Smart prefetching so the next sentence is ready before you need it
Stack:
🐍 FastAPI + Python 3.12
🔊 Supertonic-3 (ONNX Runtime inference)
📚 hashtag#Calibre for EPUB → text conversion
☁️ Deployed on hashtag#AWS hashtag#Lightsail (2 vCPUs / 2 GB RAM)
Full source on GitHub https://github.com/sharadcodes/supertonic3-book-reader
Sharad Raj (He/Him)
Posted on
Supertonic 3 + EPUB = 😍
For further actions, you may consider blocking this person and/or reporting abuse
Top comments (0)