Import
Bring in a DRM-free EPUB. Metadata, cover art, navigation, and readable sections are extracted automatically.
Local-first audiobook production
Your books.
Beautifully spoken.
Turn DRM-free EPUBs into chapterized audiobooks with a workflow made for long-form listening—not one-shot generation.
Self-hostedResumableOpen source
Interactive preview · Plan the book1 / 4
Narration plan
Choose what gets narrated.
Narranova found 12 sections and 27,442 words. Skip anything you don't want in the finished audiobook.
A safe simulation—no upload or TTS server required.
A production workflow, not a black box
From ebook to audiobook,
with every decision intact.
Narranova keeps the source book and narration plan untouched. Every enhancement, voice choice, and generated take remains traceable.
Direct
Choose what is spoken, set pauses and pronunciations, then pair the book with a reusable narrator voice.
Produce
Generate in resilient chunks. Pause safely, listen while later chapters run, and replace only the takes that need work.
Finish
Build one chapterized M4B with cover art, metadata, and markers—then reclaim the lossless working space.
Made for the long haul
Audiobooks take time.
Narranova remembers where you were.
Long-running work survives restarts, interruptions, and second thoughts. Inspect the production at the level that matters—from the whole book down to a single audio chunk.
Explore the sourceResume with confidence
Reliable pause, resume, retry, and recovery around every generation job.
Clone a narrator voice
Upload clean reference audio and save the tested recording, instruction, language, and connection as a reusable voice pair.
Review every take
Listen to completed chunks as they arrive. Delete or regenerate one without restarting the book.
Speak names correctly
Add deterministic IPA pronunciations, text normalization, headings, and scene-break pauses.
Keep storage sensible
Work in compact lossless FLAC, export once to M4B, then remove editable masters when approved.
Run it on your own machine
Your library stays yours.
Narranova runs beside an OpenMOSS text-to-speech server. Docker Compose packages the app, FFmpeg, health checks, and persistent storage into one straightforward setup.
Docker Compose OpenMOSS endpoint DRM-free EPUB
Quick start
$ git clone https://github.com/huseynli/narranova.git
$ cd narranova
$ docker compose up --detach --build Open localhost:8787