Self-hosted · open source · MIT
Read and listen to the same book, in sync.
Prosody is a self-hosted app for DRM-free books you own. It aligns each audiobook to its ebook sentence by sentence on your own machine, and keeps one place per book across your devices.
Moby-Dick · Chapter 1: Loomings
0:00 of 1:17
Call me Ishmael. Some years ago—never mind how long precisely—having little or no money in my purse, and nothing particular to interest me on shore, I thought I would sail about a little and see the watery part of the world. It is a way I have of driving off the spleen and regulating the circulation. Whenever I find myself growing grim about the mouth; whenever it is a damp, drizzly November in my soul; whenever I find myself involuntarily pausing before coffin warehouses, and bringing up the rear of every funeral I meet; and especially whenever my hypos get such an upper hand of me, that it requires a strong moral principle to prevent me from deliberately stepping into the street, and methodically knocking people’s hats off—then, I account it high time to get to sea as soon as I can. This is my substitute for pistol and ball. With a philosophical flourish Cato throws himself upon his sword; I quietly take to the ship. There is nothing surprising in this. If they but knew it, almost all men in their degree, some time or other, cherish very nearly the same feelings towards the ocean with me.
Tap any sentence to jump straight to that point in the audio.
What it does
- Text and audio in step
- When a book has both an EPUB and an M4B audiobook, Prosody aligns them sentence by sentence. The current sentence highlights as it plays, and tapping any sentence jumps the audio there.
- One place across devices
- Your place follows you between devices as a sentence, a timestamp or a page. Pause an audiobook on your phone, and your laptop opens the ebook at that same sentence.
- Built for a phone browser
- Install it to your home screen, control playback from the lock screen, open books offline and keep whole reading copies on the device when you need them.
- Works with single editions too
- Not every book needs a pair. EPUBs open to read, M4B audiobooks play by chapter, and PDFs open a page at a time.
How alignment works
An offline speech model listens to the audiobook once and maps every sentence of the text onto the audio. A second model then checks a spread of aligned sentences against the page; if too few match, the alignment is rejected rather than shown.
- 19×
- faster than real time across 72 hours of audio, on 8 CPU cores
- ~0.1 s
- median sentence start error across five measured books
- Up to 40
- sentences checked by a second model before a sync is accepted
Built to stay yours
- Plain folders, untouched
- Point Prosody at folders organised by author and title. It never moves, renames or modifies your book files; reading copies live in a separate folder.
- Offline speech models
- Both speech models are baked into the container image and run on your CPU. Your books and audio never leave your machine.
- Tailscale instead of accounts
- There are no logins or passwords. Only untagged Tailscale devices belonging to the host’s owner are admitted, identified by device name.
- One TypeScript codebase
- Licensed under MIT and structured so you, or a coding agent, can adapt its assumptions to your own hardware.
What you need
- A Linux host running Docker in swarm mode (benchmarked on 8 CPU cores, no GPU required).
- Tailscale on the host and on every device you read or listen with.
- Caddy as the reverse proxy, with a domain name pointing to the host’s tailnet address.
- DRM-free books (EPUB, M4B or PDF) in folders organised by author and title.