A free, open-source desktop app
EPUB to Audiobook
Turns any e-book into an audiobook on your own computer, then highlights each sentence on screen as it is read.
English · Mandarin (Mainland and Taiwan) · Cantonese
Contents, as the app found them
What I Talk About When I Talk About Running
- 1Contents95 w
- 2Foreword751 w
- 3Chapter One5,326 w · 3 parts
- 4Chapter Two6,294 w · 4 parts
- 5Chapter Three5,281 w · 3 parts
- 6Chapter Four4,817 w · 3 parts
- 7Chapter Five3,679 w · 2 parts
- 8Chapter Six4,990 w · 3 parts
- 9Chapter Seven3,133 w · 2 parts
- 10Chapter Eight4,002 w · 2 parts
- 11Chapter Nine5,832 w · 3 parts
- 12Afterword1,221 w
- 13Also by Haruki Murakami62 w
- 14Copyright163 w
~45,646 words~5h 4m estimated~12 min per part
Retyped from a --list dry run of the V6 command-line tool on an EPUB of Haruki Murakami's memoir (the book is not in the repo). The full output is in the write-up.
How it works
- Open an EPUB. The app finds the chapters and flags contents, copyright and "Also by" pages so you can untick them.
- Pick a voice and language. Microsoft's free edge-tts neural voices, no API key. Chinese books get a Mandarin voice automatically; switch to a Hong Kong voice for Cantonese.
- Convert. Each chapter renders to its own MP3, written safely so an interrupted run resumes where it stopped.
- Listen and read along. The built-in player highlights each sentence using the timing edge-tts sends with the audio. Or export a single
.m4bwith chapter markers.
Six versions
- V1
- Extraction, chapter detection, synthesis, atomic writes, resume
- V2
- Shared engine, timing sidecars for read-along, ID3 tags and
.m4b - V3–4
- Read-along timed directly from edge-tts boundary events, speed control, the modern interface
- V5
- Mandarin and Cantonese voices, Chinese sentence splitting and fonts
- V6
- Finds ffmpeg off PATH, so speed control and
.m4bwork on a fresh machine
About the books. The app ships no books or audio. Convert e-books you own or public-domain titles, for personal listening. The text is sent to Microsoft's edge-tts service to be voiced; everything else runs on your computer.