FR
live

Narratarr brings the Sonarr formula to audiobooks with its own built-in player

On October 29, 2026, noted.lol covered Narratarr, a self-hosted audiobook manager that pairs discovery, downloading, and playback in a single Docker container. For the self-hoster who listens to books, it is one more piece of the *arr stack — worth running on a trusted network, alongside Audiobookshelf rather than as a replacement for it.

A pair of headphones resting on a stack of old books in a dim room, a single amber button lit on one earcup.

October 29, 2026. noted.lol covers Narratarr, a new self-hosted audiobook manager shipped as a Docker image. Sonarr and Radarr have organized series and films for years; Narratarr ports the same mechanics to audiobooks. 8484 — the default port for its web interface. The point that matters: discovery, download, and playback of an audiobook now live in a single container, with no chain of services to assemble.

Discovery before download

The first thing Narratarr changes is how you enter your library. Where the classic *arr tools assume you already know what you want, Narratarr offers a genuine discovery interface. You can browse by genre or search for a specific title, with cover art and a description on hand to help you decide — an experience built for someone who browses a collection without a fixed idea in mind.

The main source is free. Narratarr connects to the LibriVox and Old-Time Radio collections on the Internet Archive, thousands of public-domain recordings. The inclusion of Old-Time Radio, those vintage radio dramas, gives the catalog a charm that commercial release lists lack. An optional feature corrects typos in search — it leans on Apple’s public API, so it is not fully local, but the application works without enabling it.

The gap audiobooks left in the *arr ecosystem

It helps to place Narratarr in a longer story. The *arr family grew in concentric circles: Sonarr for series, Radarr for films, then Lidarr for music and Readarr for books. Audiobooks fell through the cracks, stranded between two worlds — they are books, but they are consumed like music. Audiobookshelf, the heavyweight of the field, picked its side: it excels at serving and playing a library you already own, but it leaves discovery and acquisition to other tools.

That is exactly the gap Narratarr fills. Where Audiobookshelf is a playback server, Narratarr is an acquisition manager: it starts from the catalog — LibriVox, Old-Time Radio, then tomorrow the indexers through Prowlarr — to fill the library that Audiobookshelf will serve. The two are not competitors; they are complements: one finds and files, the other streams and syncs listening position across devices.

That complementarity has a practical consequence. A listener who swears by Audiobookshelf has no reason to migrate; they do have a good reason to add Narratarr upstream to automate the tedious part — finding a book, downloading it, organizing it. And because Narratarr can drop files into a tree Audiobookshelf recognizes and trigger a scan, the loop closes with no manual step.

The part any *arr user recognizes

Anyone who has run Sonarr or Radarr will recognize the logic: find a title, add it to the queue, let the application sort out the files. Downloads from LibriVox and Old-Time Radio go out directly over HTTPS, with no paid subscription. Narratarr can also talk to Prowlarr to query configured indexers, and hand Usenet downloads to SABnzbd. On the torrent side, the only current path runs through Real-Debrid, a paid service — there is no direct qBittorrent or Transmission integration yet.

The filing is clean: finished books are organized into author then title folders, with cover art and metadata saved alongside the audio files. It automates the tedious manual cleanup nobody enjoys. None of this is required to use the free collections; it is convenience, not a prerequisite.

A built-in player, no extra app

The break from other *arr tools is here: Narratarr ships its own browser-based player. It handles chapter navigation, adjustable playback speed, and resume from where you stopped — essential when a book runs dozens of hours. A sleep timer can even stop at the end of a chapter, a thoughtful touch for anyone who listens before bed without wanting to wake three hours deeper into the story.

The player treats books split across multiple MP3 files the same as M4B audiobooks, stitching them into one continuous listen. It is reachable from a phone through the browser, with no dedicated mobile app. And if you already run Audiobookshelf, there is no need to abandon it: Narratarr can drop books into a library Audiobookshelf recognizes and trigger a scan after each download.

A very young project, keep it at home

On the deployment side, the developer ships a Docker Compose configuration with AMD64 and ARM64 images. Everything runs in a single container, with two separate mounts — one for configuration, one for the library. The web interface listens on port 8484 by default. Password protection is optional, but the documentation recommends keeping Narratarr on a trusted home network rather than exposing it directly to the internet.

This is a very new project: the GitHub repository is still small, and there is little long-term feedback from real deployments. Rough edges are to be expected as development continues. The risk is not technical — it is the risk of any young software, namely a lack of track record on stability and image security before a community has stress-tested it.

In practice, that calls for two precautions. The first is to not expose Narratarr to the internet: the password option is a convenience, not a boundary, and the project lacks the security track record of an Audiobookshelf. The second is to watch the releases on the repository and rebuild the container on every update, especially when a security fix is included. A library manager that downloads and files content is an attack surface like any other — the project’s youth does not excuse it, it compounds it.

bash
# Minimal startup: clone the repository, then launch the container
git clone https://github.com/narratarr/narratarr.git
cd narratarr
docker compose up -d
# The web interface then listens on http://localhost:8484

Verdict

*If you listen to audiobooks and already run an arr stack, install Narratarr as a complement: discovery across the LibriVox and Old-Time Radio collections, paired with the built-in player, beats stitching together a pile of separate services — as long as you keep it on your local network and point it at Audiobookshelf if you already have one. If your library lives entirely in Audiobookshelf, migrate nothing: Narratarr is too young to replace a proven tool, but it can feed it. And if you have neither a trusted network nor the habit of patching your containers, stay with a cloud option — a self-hosted player is only as good as the discipline of whoever maintains it. Narratarr’s real promise is not to replace anything: it is to finally make self-hosted audiobooks as smooth as a series in Sonarr.

References

The cyber brief, every Tuesday

The flaws that matter and the patches to apply, in a ten-minute read.

No spam. One-click unsubscribe.
read next

On the same topic

selfh.st’s 2026 survey puts numbers on the Portainer exodus: 66% to 33% in a year

selfh.st’s annual survey, published November 12, 2026 from 6,221 responses (+52%), shows Portainer falling from 66% to 33% share among container-management tool users, overtaken by the Dockhand, Komodo, and Arcane wave, while 60% of respondents already use AI. If your homelab runs on Portainer CE, this shift is about you.

Immich 3.4.0 brings tags to the mobile apps, two years after their web launch

On November 10, 2026 Immich shipped version 3.4.0, finally bringing tags — launched in August 2024 but confined to the web — to the mobile apps, alongside JPEG XL and passkeys. If you organized your library with tags, mobile parity is here; upgrade and set MACHINE_LEARNING_MODEL_REVISION=v2 to reclaim the ML container’s memory win.

← Back to the feed

Type at least two characters.

↑ ↓ navigate ↵ open esc dismiss