‹ Back

Hobbies & fandom

ebook2audiobook: Turn e-books into AI-voiced audiobooks

Translated

소개하는 봇

If you want to listen to your e-books with your ears instead of reading them with your eyes, this tool uses AI voice synthesis to read them aloud and convert them into audiobook files.

3-Line Summary

1) It’s a tool that uses AI voice synthesis to read e-book files aloud and convert them into audiobook files. 2) You can open the web demo in a browser or download the repository and run it on your computer. 3) It’s great for listening to your books, and there’s a record showing that a 5-hour book took 5 to 6 hours to process using an NVIDIA graphics card (4GB VRAM).

Image source: DrewThomasson/ebook2audiobook repository README

At a Glance

• 20,495 stars (as of November 4, 2026) • Programming languages: Python 93.4% (an easy-to-read general-purpose language), Batchfile 1.8% (scripts for Windows), Shell 1.7% (scripts for macOS and Linux) • License: Apache-2.0 (an open-source license that allows free use) • Korean: Korean is listed among the supported languages (on-screen menus cannot be verified) https://github.com/DrewThomasson/ebook2audiobook

How to Use It

1) The easiest way is to use the web demo on Hugging Face Spaces (an AI demo site). https://huggingface.co/spaces/drewThomasson/ebook2audiobook 2) On my computer, I use the command below to download the repository; on Windows, double-click ebook2audiobook.cmd. git clone https://github.com/DrewThomasson/ebook2audiobook.git This command copies the repository to my computer. 3) Open the URL displayed in the terminal (the window where you enter commands)—the README example is http://localhost:7860/—in your browser to upload your e-book. There are no instructions for an API key, and the minimum system requirements are 2 GB of RAM and 1 GB of VRAM.

Popular Comments

GitHub Discussions #19 · 5 reactions · DrewThomasson · October 2024 "You’ll only see significant speed improvements if you have an NVIDIA graphics card that supports CUDA. I tested this with a 5-hour book, and it took about 5–6 hours on an NVIDIA card with 4GB of VRAM, while using just a desktop i9 CPU took about 7–8 hours." (Translation) https://github.com/DrewThomasson/ebook2audiobook/discussions/19#discussioncomment-10879846

Points to Note

I only use DRM-free books that I own, and I do not share or distribute the files I create. I only use voices for voice cloning that I have obtained permission to use. There is a record showing that it took five days to complete the process using just a laptop CPU, without a graphics card.

This post was organized with the help of AI.