This is a plugin that allows Beets to manage audiobook collections.
It fetches audiobook metadata via the Audible and Audnex API. With this data, it ensures books have the correct tags and makes the collection ready to be served by Plex, Audiobookshelf or Booksonic.
seanap's audiobook organization guide describes a workflow for adding tags to audiobooks and moving the files to the right folders.
However, it relies on using Mp3tag, a gui tool which does not lend itself to automation. Mp3tag works only on Windows.
This Beets plugin solves both problems.
Install via pip: pip install beets-audible beets-copyartifacts3
(copyartifacts is optional but recommended). See the next section instead if you're running Beets in Docker (highly recommended as it makes it easier to maintain a separate Beets installation dedicated to audiobooks).
Use a separate beets config and database for managing audiobooks. This is the recommended Beets config for this plugin:
# add audible to the list of plugins
# copyartifacts is optional but recommended if you're manually specifying metadata via metadata.yml, see the "Importing non-audible content" section
# also add the "web" plugin if using the docker image
plugins: audible copyartifacts edit fromfilename scrub
directory: /audiobooks
# Place books in their own folders to be compatible with Booksonic and Audiobookshelf servers
paths:
# For books that belong to a series
"albumtype:audiobook series_name::.+ series_position::.+": $albumartist/%ifdef{series_name}/%ifdef{series_position} - $album%aunique{}/$track - $title
"albumtype:audiobook series_name::.+": $albumartist/%ifdef{series_name}/$album%aunique{}/$track - $title
# Stand-alone books
"albumtype:audiobook": $albumartist/$album%aunique{}/$track - $title
default: $albumartist/$album%aunique{}/$track - $title
singleton: Non-Album/$artist - $title
comp: Compilations/$album%aunique{}/$track - $title
albumtype_soundtrack: Soundtracks/$album/$track $title
# disables musicbrainz lookup, as it doesn't help for audiobooks
musicbrainz:
host: localhost:5123
audible:
# if the number of files in the book is the same as the number of chapters from Audible,
# attempt to match each file to an audible chapter
match_chapters: true
source_weight: 0.0 # disable the source_weight penalty
fetch_art: true # whether to retrieve cover art
include_narrator_in_artists: true # include author and narrator in artist tag. Or just author
keep_series_reference_in_title: true # set to false to remove ", Book X" from end of titles
keep_series_reference_in_subtitle: true # set to false to remove subtitle if it contains the series name and the word book ex. "Book 1 in Great Series", "Great Series, Book 1"
write_description_file: true # output desc.txt
write_reader_file: true # output reader.txt
copyartifacts:
extensions: .yml # so that metadata.yml is copied, see below
scrub:
auto: yes # optional, enabling this is personal preference
Run the beet --version
command and verify that the audible plugin appears in the list of plugins.
Create the following folder structure:
beets
config/
scripts/
install-deps.sh # see step 3
docker-compose.yml # see step 2
Save the following as the docker-compose file:
---
version: "3"
services:
beets:
image: lscr.io/linuxserver/beets:1.6.0-ls180
container_name: beets
environment:
# Update as needed
- PUID=1000
- PGID=1000
- TZ=Asia/Singapore
volumes:
- ./config:/config
- ./scripts:/custom-cont-init.d
- /path/to/audiobooks:/audiobooks
- /path/to/import/books/from:/input
restart: unless-stopped
Save the following under scripts/install-deps.sh
:
#!/bin/bash
echo "Installing dependencies..."
# copyartifacts is optional but recommended
pip install --no-cache-dir beets-audible beets-copyartifacts3
Spin up the container: docker-compose up -d
Update the config in config/config.yaml
as described above.
In the docker container, run the beet --version
command and verify that the audible plugin appears in the list of plugins.
When importing audiobooks into Beets, ensure that the files for each book are in its own folder, even if the audiobook only consists of a single file. This is so that the files for a book are treated as an album by Beets. Avoid putting files from multiple books in the same folder.
When ready, start the import by executing the following command in the container:
beet import /path/to/audiobooks
The following sources of information are used to search for book matches in order of preference:
metadata.yml
(see below)If you're not getting a match for a book, try the following:
E
when Beets prompts you about not being able to find a match. This prompts for the artist and album name. If the wrong book is being matched because there are other books with similar names on Audible, try using the audiobook's asin as the artist and title as the album.metadata.yml
if it isn't on Audible (see the next section).The plugin gets chapter data of each book and tries to match them to the imported files if and only if the number of imported files is the same as the number of chapters from Audible. This can fail and cause inaccurate track assignments if the lengths of the files don't match Audible's chapter data. If this happens, set the config option match_chapters
to false
temporarily and try again, and remember to uncomment that line once done.
The plugin can search Goodreads to find the original publication date of the work the audiobook is based on by searching on the ASIN. To enable this option you need a Goodreads API key and you must set that key in the audible plugin config
audible:
goodreads_apikey: [APIKEYHERE] #optional
If you want this date used as the release date for the audiobook, you must set original_date to yes in your beets config
The plugin looks for a file called metadata.yml
in each book's folder during import. If this file is present, it exclusively uses the info in it for tagging and skips the Audible lookup.
This is meant for importing audio content that isn't on Audible.
Here's an example of what metadata.yml
should look like:
---
# These are all required fields
title: The Lord of the Rings (BBC Dramatization)
authors: ["J. R. R. Tolkien", "Brian Sibley", "Michael Bakewell"]
narrators:
- "Ian Holm (as Frodo)"
- "Sir Michael Hordern (as Gandalf)"
- "Robert Stephens (as Aragorn)"
- "John Le Mesurier"
description: |
This audio set includes: The Fellowship of the Ring; The Two Towers; and The Return of the King.
Undertaking the adaptation of J.R.R. Tolkien's best-known work was an enormous task, but with its first broadcast on BBC Radio 4 on March 8, 1981, this dramatized tale of Middle Earth became an instant global classic. Thrilling dramatization by Brian Sibley and Michael Bakewell it boasts a truly outstanding cast including Ian Holm (as Frodo), Sir Michael Hordern (as Gandalf), Robert Stephens (as Aragorn), and John Le Mesurier. Tolkiens tale relates the perilous attempt by Frodo Baggins and company to defeat the evil Saruman and dispose of the Ruling Ring. Brian Sibley wrote the opening and closing narration for the character of Frodo, played by Ian Holm, who now stars as Bilbo in the feature films based on The Lord of the Rings.
genres: ["fantasy"]
releaseDate: 2008-08-19
publisher: BBC Audiobooks
# optional fields
language: English # defaults to "English" if not specified
subtitle: "some subtitle"
series: The Lord Of The Rings
seriesPosition: "1-3"
The copyartifacts plugin ensures that metadata.yml
is copied over during the import, as it gets left in the source folder otherwise.
The config above places books according to this folder structure, which can be changed by editing the path config.
Terry Goodkind/
Sword of Truth/
1 - Wizards First Rule/
cover.png
desc.txt
reader.txt
wizards first rule.m4b
George Orwell/
Animal Farm/
Animal Farm.m4b
cover.png
desc.txt
reader.txt
Desc.txt and reader.txt contain the book description and narrator populated from Audible.
The plugin writes the following tags:
ID3 Tag | Audible.com Value |
---|---|
TIT1 (CONTENTGROUP) |
Series, Book # |
TALB (ALBUM) |
Title |
TIT3 (SUBTITLE) |
Subtitle |
TPE1 (ARTIST) |
Author, Narrator |
TPE2 (ALBUMARTIST) |
Author |
TCOM (COMPOSER) |
Narrator |
TCON (GENRE) |
Genre1/Genre2 |
TDRC and TDRL (release date) |
audio publication date |
COMM or desc for m4b files (COMMENT) |
Publisher's Summary |
TSOA (ALBUMSORT) |
If ALBUM only, then %Title% If ALBUM and SUBTITLE, then %Title% - %Subtitle% If Series, then %Series% %Series-part% - %Title% |
TPUB (PUBLISHER) |
Publisher |
ASIN (ASIN) |
Amazon Standard Identification Number |
stik (media type), m4b only |
2 (audiobook) |
shwm (show movement), m4b only |
1 if part of a series |
MVNM (MOVEMENTNAME) |
Series |
MVIN / MVI for m4b files (MOVEMENT) |
Series Book # |
TXXX_SERIES (SERIES) |
Series |
TXXX_SERIES-PART |
Series position |
If the directory where Beets imports audiobooks to is also where you've set Plex to serve content from, you can enable the plexupdate plugin to notify Plex when new books are imported.