Turn any book or text into an audiobook, generated on your own machine. Clone your own voice, fine-tune the read, then master and export to MP3 or M4B. Runs OFFLINE, no subscription, no cloud.

Sign in to add this item to your wishlist, follow it, or mark it as ignored

Notice: Narration OS is no longer available on the Steam store.
This game is not yet available on Steam

Coming soon

Interested?
Add to your wishlist and get notified when it becomes available.
 

About This Software

NarrationOS turns any book or any text into an audiobook, generated entirely on your own machine. Drop in a file, pick a voice, and get back a finished narration you can listen to, master, and keep. NO subscription, NO uploads, NO cloud, completely OFFLINE. Your books, performed.

Built by a solo developer who couldn't wait for an audiobook that never came, NarrationOS runs the whole pipeline locally on your GPU, powered by Resemble AI's Chatterbox voice model.

Bring any book

  • Import EPUB, PDF, DOCX, RTF, TXT, Markdown, and HTML, automatically split into chapters.

  • No file? Paste any text, give it a title, and narrate it in seconds.

  • Scanned PDFs are handled with built-in OCR.

Your own voice

  • Clone a voice from a short 10 to 15 second recording, then use it across every book.

  • Cloning is consent-gated: you confirm your right to the voice, and a receipt is signed locally with your own key.

  • Preview any voice before you commit it to a book.

Shape the performance

  • The Narration Editor lets you fine-tune a chapter before you render: set book-wide name pronunciations and audition individual lines.

  • Detect sound effects and laughter so they're voiced as real sounds instead of spelled out.

  • Every change is previewed in the text before you commit. Nothing touches the audio until you press Render.

Listen your way

  • A karaoke reader highlights each word as it's read. Click any line to jump there.

  • Switch to a cinematic teleprompter layout for a focused, immersive read.

  • Bookmarks, adjustable speed, a sleep timer, and full keyboard control.

Finish and own it

  • Master your audiobook with target loudness, a clarity EQ, and a de-esser.

  • Export to MP3, WAV, or chaptered M4B. Real files on your drive that you keep forever.

Runs on your hardware

  • Best on NVIDIA (CUDA) GPUs, the primary and most-tested path. AMD (RDNA3/4, ROCm) are also fully GPU-accelerated. CPU-only works too, just slower.

  • The AI runtime is bundled, so no internet is needed after install. 

  • Runs Completely OFFLINE, NO cloud, NO subscription.

An honest word from the developer

NarrationOS is a solo project and still in active development, so I'd rather be upfront than oversell it. It already does the thing I built it for: it let me finally listen to the books I wanted, in the voices I wanted, without waiting on anyone. I'll keep improving it.

What's included

  • The full app, version 1.0.0. + future updates will be included in this purchase.

  • No stock voices. You clone your own from a short recording during setup, so the voices on your book is of your choice.

  • A sample book: The Lighthouse Keeper. An original short story written for NarrationOS, already in your library on first launch, with named characters and dialogue so you can try every feature on a real book.

AI Generated Content Disclosure

The developers describe how their game uses AI Generated Content like this:

NarrationOS uses generative AI to create its output. It synthesizes speech from your books using an AI voice model (Resemble AI's Chatterbox) and can generate a custom voice from a short recording you supply. A local AI language model assists with text analysis. All generation happens locally on your device. You are responsible for having the rights to any content or voice you use, and voice cloning requires your confirmation of consent.

System Requirements

    Minimum:
    • OS: Windows 10 64-bit (version 1803 or newer)
    • Processor: Quad-core x64 CPU (Intel Core i5 / AMD Ryzen 5 or better)
    • Memory: 8 GB RAM
    • Graphics: Runs on CPU without a GPU, but far slower than realtime.
    • Storage: 40 GB available space
    • Sound Card: Any Windows-compatible audio output
    • Additional Notes: First launch installs the bundled AI runtime (~7 GB, one-time, no internet needed). A GPU is strongly recommended — NVIDIA (CUDA) runs best; on CPU-only, rendering is much slower than realtime.
    Recommended:
    • OS: Windows 11 64-bit
    • Processor: 8-core x64 CPU (Intel Core i7 / AMD Ryzen 7 or better)
    • Memory: 16 GB RAM
    • Graphics: 12 GB+ VRAM GPU. Best on NVIDIA RTX 3060 12GB / 4060TI or newer (CUDA). Also supported: AMD Radeon RX 7000/9000 (RDNA3/4, ROCm)
    • Storage: 60 GB available space
    • Sound Card: Any Windows-compatible audio output
    • Additional Notes: NarrationOS runs best on NVIDIA (CUDA) GPUs — the primary, most-tested path. AMD (RDNA3/4, ROCm) are fully GPU-accelerated and supported. Any other GPU or CPU still works, just slower.
Review Filters