AUDIBLE-CONVERT

Audible AAX/AAXC conversion with emphasis on:

• automated retreival
• audio quality
• single file
• proper tagging and chaptering
• directory structure
• hands off functionality (or as much as possible)

2024.03.04

Have done a boat load of testing and various modifications to try and get the process automatable. I have succeeded in automation, however whatever Audible is doing for naming conventions is all over the map. This makes nailing down a "one size fits all" solution nearly impossible. For instance, a single author could be spelled differently for each book in the Audible library...

• Jon Doe
• Jon T. Doe
• Jon T Doe
• Dr. Jon T. Doe
• Dr Jon T Doe

You get the picture. What makes this even more fun to deal with is that any of the book information could change at any time. Using the audible-cli from mkb79, I pulled down a library.tsv and had been working through some metadata configurations, but something was off. It took me a little bit to figure out that during the period of time when I pulled the library.tsv and then pulled another one to see if I screwed something up (about an hour later), a couple of the books I was working with had its information changed. This has happened a few times already. So I have resolved to the fact that I need a single source of truth, and something that I can verify myself.

That said, I have incorporated the ability to use a custom.tsv and have split up a function in the ac.run so that there can be a "initial" run, and then moving forward the script can be schedule to run. The idea behind this is that I can pull down my entire library and generate a library.tsv file. Then I can go through the library.tsv and manually correct any of the information and make changes how I see fit for my library. Then I can save this file off as a custom.tsv, and run the book conversions against my own vetted information.

Scheduling the script afterwards is not such a deal breaker, because I have also included a check for the last time the script had run. This check allows the script to only look for anything new (also an ability worked into the audible-cli tool). By dropping any new books into a "new" directory, makes them much easier to vet before I incorporate them into my main library.

So, whats been done:

• ac.core: defaults have been set within the file to create a single m4a file with chapter information

• ac.core: uses the library.tsv file for <title>, <subtitle>, <artist>, <composer>, <series>, and <series_sequence>

• ac.core: uses mediainfo for <publisher> and <description>

• ac.core: ffprobe and audible-cli provide any remaining tag and chapter information

• ac.core: variable checks have been implemented to manipulate how a "series" book should be named verses a "standalone"

• ac.core: the asin is now being plugged into the <comment> tag

• ac.run: this script was created to take care of how ac.core is called and handled

• ac.run: function for pulling down the entire library and creating the library.tsv file

• ac.run: function for creating and checking a timestamp of a previous run when downloading new books

• ac.run: function for checking for a custom.tsv before kicking off a conversion

2024.02.29

HAPPY LEAP YEAR!

2024.02.26

• The current naming scheme of my library uses <author> / <book title> / <book title>.<ext>. This may be fairly common of a layout, but when it comes to "book title" well, there are many ways this could be formatted. The majority of my collection is comprised of "series" based books. But there are also single standalone books intermingling in there as well. So for example, I may have a author folder with series and standalone books,

  📁 Keith C. Blackmore
    📁 Mountain Man, Book 01 - Mountain Man
     🔊 Book 01 - Mountain Man.m4a
    📁 Mountain Man, Book 02 - Safari
     🔊 Book 02 - Safari.m4a
    📁 Mountain Man, Book 03 - Hellifax
     🔊 Book 03 - Hellifax.m4a
    📁 The Missing Boatman
     🔊 The Missing Boatman.m4a
    📁 White Sands, Red Steel
     🔊 White Sands, Red Steel.m4a

  In order to facilitate this layout, some tweaks to the script were required.

  The script already had code support for mkb79's audible-cli. This support allowed the script to obtain "series" information. The "audible-cli" tool is a CLI for mkb79's other package Audible. This tool sets up the authentication and registration to Audible, but also uses the Audible API to collect book information.
  Being that the script is already setup to use the "Audible" package book information (library.tsv) for series data, I have added additional calls to also extract the "title" and "subtitle" data. With this information I was able to manipulate the folder structure and metadata.

  • Series books now get: <series_title>, <series_sequence> - <title>
  • Standalone books get: <title> - <subtitle>

  Included are checks to make sure that "subtitle" has a value, and if it does not, then it will not be used.
  These variants have been used to re-create the <album> metadata as well, so that proper tagging is also a thing. In addition zero padding has been incorporated, so that "series_sequence" will start with a 0 on single digit series titles (e.g. "Book 01", "Book 02", etc).

• Audible also uses colons (:) in the book names (all my series titled books had colons). This required some additional tweaks. Viewing the folder structure of books with colons on a windows machine was just not working out.

• The script used a bash noclobber option that just wasn't working very well. Or let me say, it worked, but on certain book titles instead of skipping the book, it would create the book folder structure again in the current book folder and then re-rip the book into the new folder. If I ran the script again, it would do this again and again, creating several rips of the book, several tiers deep.
  I have pulled the "noclobber" option out, and replaced it with a overwrite=0 variable, and some "if var exists, then..." code.

• Renamed script to audible-convert
  The script can convert to more than just MP3, so it should have a bit more universal name.

2024.02.25

• fresh start

Warning

Anti-Piracy Notice

Note that this project does NOT "crack" the DRM. It simply allows the user to use their own encryption key (fetched from Audible servers) to decrypt the audiobook in the same manner that the official audiobook playing software does.

Please only use this application for gaining full access to your own audiobooks for archiving/conversion/convenience. DeDRMed audiobooks should not be uploaded to open servers, torrents, or other methods of mass distribution. No help will be given to people doing such things. Authors, retailers, and publishers all need to make a living, so that they can continue to produce audiobooks for us to hear, and enjoy. Don’t be a parasite.

_{Borrowed from original documentation, who borrowed from Apprentice Alf’s Blog.}

Note

This is the script (was the script) from the now defunct AAXtoMP3. Some really good work was thrown at that project. I have however made modifications to better suit my requirements.

The original AAXtoMP3 documentation for reference is below.

AAXtoMP3 Documentation

AAXtoMP3

The purpose of this software is to convert AAX (or AAXC) files to common MP3, M4A, M4B, flac and ogg formats through a basic bash script frontend to FFMPEG.

Audible uses the AAX file format to maintain DRM restrictions on their audio books and if you download your book through your library it will be stored in this format.

The purpose of this software is not to circumvent the DRM restrictions for audio books that you do not own in that you do not have them on your personal Audible account. The purpose of this software is to create a method for you to download and store your books just in case Audible fails for some reason.

Requirements

• bash 3.2.57 or later tested
• ffmpeg version 2.8.3 or later (4.4 or later if the input file is .aaxc)
• libmp3lame - (typically 'lame' in your system's package manager)
• GNU grep - macOS or BSD users may need to install through package manager
• GNU sed - see above
• GNU find - see above
• jq - only if --use-audible-cli-data is set or if converting an .aaxc file
• mp4art used to add cover art to m4a and m4b files. Optional
• mediainfo used to add additional media tags like narrator. Optional

Usage(s)

bash AAXtoMP3 [-f|--flac] [-o|--opus] [-a|--aac] [-s|--single] [--level <COMPRESSIONLEVEL>] [-c|--chaptered] [-e:mp3] [-e:m4a] [-e:m4b] [-A|--authcode <AUTHCODE>] [-n|--no-clobber] [-t|--target_dir <PATH>] [-C|--complete_dir <PATH>] [-V|--validate] [--use-audible-cli-data]] [-d|--debug] [-h|--help] [--continue <CHAPTERNUMBER>] <AAX/AAXC INPUT_FILES>...

or if you want to get guided through the options

bash interactiveAAXtoMP3 [-a|--advanced] [-h|--help]

• <AAX INPUT_FILES>... are considered input file(s), useful for batching!

Options for AAXtoMP3

• -f or --flac Flac Encoding and as default produces a single file.
• -o or --opus Ogg/Opus Encoding defaults to multiple file output by chapter. The extension is .ogg
• -a or --aac AAC Encoding and produce a m4a single files output.
• -A or --authcode <AUTHCODE> for this execution of the command use the provided <AUTHCODE> to decode the AAX file. Not needed if the source file is .aaxc.
• -n or --no-clobber If set and the target directory already exists, AAXtoMP3 will exit without overwriting anything.
• -t or --target_dir <PATH> change the default output location to the named <PATH>. Note the default location is ./Audiobook of the directory to which each AAX file resides.
• -C or --complete_dir <PATH> a directory to place aax files after they have been decoded successfully. Note make a back up of your aax files prior to using this option. Just in case something goes wrong.
• -V or --validate Perform 2 validation tests on the supplied aax files. This is more extensive than the normal validation as we attempt to transcode the aax file to a null file. This can take a long period of time. However it is useful when inspecting a large set of aax files prior to transcoding. As download errors are common with Audible servers.
• -e:mp3 Identical to defaults.
• -e:m4a Create a m4a audio file. This is identical to --aac
• -e:m4b Create a m4b audio file. This is the book version of the m4a format.
• -s or --single Output a single file for the entire book. If you only want a single ogg file for instance.
• -c or --chaptered Output a single file per chapter. The --chaptered will only work if it follows the --aac -e:m4a -e:m4b --flac options.
• --continue <CHAPTERNUMBER> If the splitting into chapters gets interrupted (e.g. by a weak battery on your laptop) you can go on where the process got interrupted. Just delete the last chapter (which was incompletely generated) and redo the task with "--continue <CHAPTERNUMBER>" where CHAPTERNUMBER is the chapter that got interrupted.
• --level <COMPRESSIONLEVEL> Set compression level. May be given for mp3, flac and opus.
• --keep-author <FIELD> If a book has multiple authors and you don't want all of them in the metadata, with this flag you can specify a specific author (1 is the first, 2 is the second...) to keep while discarding the others.
• --author <AUTHOR> Manually set the author metadata field, useful if you have multiple books of the same author but the name reported is different (eg. spacing, accents..). Has precedence over --keep-author.
• -l or --loglevel <LOGLEVEL> Set loglevel: 0 = progress only, 1 (default) = more information, output of chapter splitting progress is limitted to a progressbar, 2 = more information, especially on chapter splitting, 3 = debug mode
• --dir-naming-scheme <STRING> or -D Use a custom directory naming scheme, with variables. See below for more info.
• --file-naming-scheme <STRING> or -F Use a custom file naming scheme, with variables. See below for more info.
• --chapter-naming-scheme <STRING> Use a custom chapter naming scheme, with variables. See below for more info.
• --use-audible-cli-data Use additional data got with mkb79/audible-cli. See below for more info. Needed for the files in the aaxc format.
• --audible-cli-library-file or -L Path of the library-file, generated by mkb79/audible-cli (audible library export -o ./library.tsv). Only available if --use-audible-cli-data is set. This file is required to parse additional metadata such as $series or $series_sequence.
• --ffmpeg-path Set the ffmpeg/ffprobe binaries folder. Both of them must be executable and in the same folder.
• --ffmpeg-name Set a custom name for the ffmpeg binary. Must be executable and in path, or in custom path specified by --ffmpeg-path.
• --ffprobe-name Set a custom name for the ffprobe binary. Must be executable and in path, or in custom path specified by --ffmpeg-path.

Options for interactiveAAXtoMP3

• -a or --advanced Get more options to choose. Not used right now.
• -h or --help Get a help prompt. This script presents you the options you chose last time as default. When you get asked for the aax-file you may just drag'n'drop it to the terminal.

AUTHCODE

Your Audible auth code (it won't correctly decode otherwise) (not required to decode the aaxc format).

Determining your own AUTHCODE

You will need your authentication code that comes from Audible's servers. This will be used by ffmpeg to perform the initial audio convert. You can obtain this string from a tool like audible-activator or like audible-cli.

Specifying the AUTHCODE.

In order of precidence.

1. --authcode [AUTHCODE] The command line option. With the highest precedence.
2. .authcode If this file is placed in the current working directory and contains only the authcode it is used if the above is not.
3. ~/.authcode a global config file for all the tools. And is used as the default if none of the above are specified. Note: At least one of the above must be exist if converting aax files. The code must also match the encoding for the user that owns the AAX file(s). If the authcode does not match the AAX file no transcoding will occur.

MP3 Encoding

• This is the default encoding
• Produces 1 or more mp3 files for the AAX title.
• The default mode is chaptered
• If you want a mp3 file per chapter do not use the --single option.
• A m3u playlist file will also be created in this instance in the case of default chaptered output.
• --level has to be in range 0-9, where 9 is fastest and 0 is highest quality. Please note: The quality can never become higher than the qualitiy of the original aax file!

Ogg/Opus Encoding

• Can be done by using the -o or --opus command line switches
• The default mode is chaptered
• Opus coded files are stored in the ogg container format for better compatibility.
• --level has to be in range 0-10, where 0 is fastest and 10 is highest quality. Please note: The quality can never become higher than the qualitiy of the original aax file!

AAC Encoding

• Can be done by using the -a or --aac command line switches
• The default mode is single
• Designed to be the successor of the MP3 format
• Generally achieves better sound quality than MP3 at the same bit rate.
• This will only produce 1 audio file as output.

FLAC Encoding

• Can be done by using the -f or --flac command line switches
• The default mode is single
• FLAC is an open format with royalty-free licensing
• This will only produce 1 audio file as output. If you want a flac file per chapter do use -c or --chaptered.
• --level has to be in range 0-12, where 0 is fastest and 12 is highest compression. Since flac is lossless, the quality always remains the same.

M4A and M4B Containers

• These containers were created by Apple Inc. They were meant to be the successor to mp3.
• M4A is a container that is meant to hold music and is typically of a higher bitrate.
• M4B is a container that is meant to hold audiobooks and is typically has bitrates of 64k and 32k.
• Both formats are chaptered
• Both support coverart internal
• The default mode is single

Validating AAX files

• The --validate option will result in only a validation pass over the supplied aax file(s). No transcoding will occur. This is useful when you wish to ensure you have a proper download of your personal Audible audio books. With this option all supplied books are validated.
• If you do NOT supply the --validate option all audio books are still validated when they are processed. However if there is an invalid audio book in the supplied list of books the processing will stop at that point.
• A third test is performed on the file where the entire file is inspected to see if it is valid. This is a lengthy process. However it will not break the script when an invalid file is found.
• The 3 test current are:
  1. aax present
  2. meta data header in file is valid and complete
  3. entire file is valid and complete. only executed with the --validate option.

Defaults

• Default out put directory is the base directory of each file listed. Plus the genre, Artist and Title of the Audio Book.
• The default codec is mp3
• The default output is by chapter.

Custom naming scheme

The following flags can modify the default naming scheme:

• --dir-naming-scheme or -D
• --file-naming-scheme or -F
• --chapter-naming-scheme

Each flag takes a string as argument. If the string contains a variable defined in the script (eg. artist, title, chapter, narrator...), the corresponding value is used. The default options correspond to the following flags:

• --dir-naming-scheme '$genre/$artist/$title'
• --file-naming-scheme '$title'
• --chapter-naming-scheme '$title-$(printf %0${#chaptercount}d $chapternum) $chapter'

Additional notes:

• If a command substitution is present in the passed string, (for example $(printf %0${#chaptercount}d $chapternum), used to pad with zeros the chapter number), the commands are executed. So you can use --dir-naming-scheme '$(date +%Y)/$artist', but using --file-naming-scheme '$(rm -rf /)' is a really bad idea. Be careful.
• You can use basic text, like --dir-naming-scheme 'Converted/$title'
• You can also use shell variables as long as you escape them properly: CustomGenre=Horror ./AAXtoMP3 --dir-naming-scheme "$CustomGenre/\$artist/\$title" *.aax
• If you want shorter chapter names, use --chapter-naming-scheme '$(printf %0${#chaptercount}d $chapternum) $chapter': only chapter number and chapter name
• If you want to append the narrator name to the title, use --dir-naming-scheme '$genre/$artist/$title-$narrator' --file-naming-scheme '$title-$narrator'
• If you don't want to have the books separated by author, use --dir-naming-scheme '$genre/$title'
• To be able to use $series or $series_sequence in the schemes the following is required:
  • --use-audible-cli-data is set
  • you have pre-generated the library-file via audible library export -o ./library.tsv
  • you have set the path to the generated library-file via --audible-cli-library-file ./library.tsv

Installing Dependencies.

In general, take a look at command-not-found.com

FFMPEG,FFPROBE

Ubuntu, Linux Mint, Debian

sudo apt-get update
sudo apt-get install ffmpeg libav-tools x264 x265 bc

In Debian-based system's repositories the ffmpeg version is often outdated. If you want to convert .aaxc files, you need at least ffmpeg 4.4. So if your installed version needs to be updated, you can either install a custom repository that has the newer version, compile ffmpeg from source or download pre-compiled binaries. You can then tell AAXtoMP3 to use the compiled binaries with the --ffmpeg-path flag. You need to specify the folder where the ffmpeg and ffprobe binaries are. Make sure they are both executable.

If you have snapd installed, you can also install a recent version of 4.4 from the edge channel:

snap install ffmpeg --edge

In this case you will need to confiure a custom path and binary name for ffprobe, --ffmpeg-path /snap/bin/ --ffprobe-name ffmpeg.ffprobe.

Fedora

Fedora users need to enable the rpm fusion repository to install ffmpeg. Version 22 and upwards are currently supported. The following command works independent of your current version:

sudo dnf install https://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-$(rpm -E %fedora).noarch.rpm https://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-$(rpm -E %fedora).noarch.rpm

Afterwards use the package manager to install ffmpeg:

sudo dnf install ffmpeg

RHEL or compatible like CentOS

RHEL version 6 and 7 are currently able to use rpm fusion. In order to use rpm fusion you have to enable EPEL, see http://fedoraproject.org/wiki/EPEL

Add the rpm fusion repositories in version 6

sudo yum localinstall --nogpgcheck https://download1.rpmfusion.org/free/el/rpmfusion-free-release-6.noarch.rpm https://download1.rpmfusion.org/nonfree/el/rpmfusion-nonfree-release-6.noarch.rpm

or version 7:

sudo yum localinstall --nogpgcheck https://download1.rpmfusion.org/free/el/rpmfusion-free-release-7.noarch.rpm https://download1.rpmfusion.org/nonfree/el/rpmfusion-nonfree-release-7.noarch.rpm

then install ffmpeg:

sudo yum install ffmpeg

MacOS

brew install ffmpeg
brew install gnu-sed
brew install grep
brew install findutils

mp4art/mp4chaps

Note: This is an optional dependency, required for adding cover art to m4a and b4b files only.

Ubuntu, Linux Mint, Debian

sudo apt-get update
sudo apt-get install mp4v2-utils

On Debian and Ubuntu the mp4v2-utils package has been deprecated and removed, as the upsteam project is no longer maintained. The package was removed in Debian Buster, and Ubuntu Focal [ 20.04 ].

CentOS, RHEL & Fedora

# CentOS/RHEL and Fedora users make sure that you have enabled atrpms repository in system. Let’s begin installing FFmpeg as per your operating system.
yum install mp4v2-utils

MacOS

brew install mp4v2

mediainfo

Note: This is an optional dependency.

Ubuntu, Linux Mint, Debian

sudo apt-get update
sudo apt-get install mediainfo

CentOS, RHEL & Fedora

yum install mediainfo

MacOS

brew install mediainfo

AAXC files

The AAXC format is a new Audible encryption format, meant to replace the old AAX. The encryption has been updated, and now to decrypt the file the authcode is not sufficient, we need two "keys" which are unique for each audiobook. Since getting those keys is not simple, for now the method used to get them is handled by the package audible-cli, that stores them in a file when downloading the aaxc file. This means that in order to decrypt the aaxc files, they must be downloaded with audible-cli. Note that you need at least ffmpeg 4.4.

Audible-cli integration

Some information are not present in the AAX file. For example the chapters's title, additional chapters division (Opening and End credits, Copyright and more). Those information are avaiable via a non-public audible API. This repo provides a python API wrapper, and the audible-cli packege makes easy to get more info. In particular the flags --cover --cover-size 1215 --chapter downloads a better-quality cover (.jpg) and detailed chapter infos (.json). More info are avaiable on the package page.

Some books might not be avaiable in the old aax format, but only in the newer aaxc format. In that case, you can use audible-cli to download them. For example, to download all the books in your library in the newer aaxc format, as well as chapters's title and an HQ cover: audible download --all --aaxc --cover --cover-size 1215 --chapter.

To make AAXtoMP3 use the additional data, specify the --use-audible-cli-data flag: it expects the cover and the chapter files (and the voucher, if converting an aaxc file) to be in the same location of the AAX file. The naming of these files must be the one set by audible-cli. When converting aaxc files, the variable is automatically set, so be sure to follow the instructions in this paragraph.

For more information on how to use the audible-cli package, check out the git page audible-cli.

Please note that right now audible-cli is in dev stage, so keep in mind that the naming scheme of the additional files, the flags syntax and other things can change without warning.

Anti-Piracy Notice

Note that this project does NOT ‘crack’ the DRM. It simply allows the user to use their own encryption key (fetched from Audible servers) to decrypt the audiobook in the same manner that the official audiobook playing software does.

Please only use this application for gaining full access to your own audiobooks for archiving/conversion/convenience. DeDRMed audiobooks should not be uploaded to open servers, torrents, or other methods of mass distribution. No help will be given to people doing such things. Authors, retailers, and publishers all need to make a living, so that they can continue to produce audiobooks for us to hear, and enjoy. Don’t be a parasite.

This blurb is borrowed from the https://apprenticealf.wordpress.com/ page.

License

Changed the license to the WTFPL, do whatever you like with this script. Ultimately it's just a front-end for ffmpeg after all.

Need Help?

I'll help out if you are having issues, just submit and issue and I'll get back to you when I can.