AudioBookConverter turns a folder of mp3 files into a single m4b

One book, not a folder of files

AudioBookConverter was originally designed as a simple and lightweight application to convert a set of mp3 files to a single m4b, which is an aac container, to be played on an i-device, or any device now.

That origin explains most of the design. The unit of work is a book, not a file. The application takes a set of tracks, works out where the chapters go, finds the artwork, writes the tags and produces one file that an Apple device will show in the Books app with a proper title, a cover and a chapter list.

Over time it evolved, and the current version can support conversion of most of the popular formats to m4b, it also can handle existing m4b files and process them, can work smart to find relevant artwork, and can do advanced chapter management.

The implementation is Java, the licence is GPL-2.0, and the last push was on 2026-10-15. The release history is patch-level and recent, with 6.6.8 on 2025-12-15 and 6.6.7 and 6.6.6 a few days earlier.

The primary language being Java with a JavaFX interface is worth noting, because it explains both the smooth native feel of the desktop application and the size of the installer.

What it reads and what it writes

The format lists are short and they are the clearest statement of the scope.

The supported source formats are mp3, flac and cue, wma, m4a and m4b, aac, and wav. The supported output formats are m4b with aac, ogg with opus, and mp3 with lame.

So the reading side is broad and the writing side is narrow, and that is the right way round for the tool. A real library is a mixture, and the interesting case is a mixed library being brought into one book.

The presence of flac and cue as an input pair is worth a note, because cue sheets are how a single flac file describes a multi-track album, and a converter that understands them can reconstruct chapter boundaries that exist nowhere else in the source. The same applies to the m4b input, since an existing m4b already carries a chapter list.

The output side tells you who the tool is for. m4b is the Apple format, and the alternatives are the two open codecs, opus in ogg and mp3 with lame. There is no wav output, which is consistent with a tool whose purpose is delivery to a device rather than archival.

The ffmpeg tag in the topic list is the giveaway about the implementation, since a converter that handles this many codecs is almost certainly delegating the actual encoding to ffmpeg rather than implementing the codecs itself.

The features that are not just a format conversion

The main features list has eight entries, and four of them are the reason to use this tool rather than a command line ffmpeg invocation.

Parallel encoding, where more cores mean faster results, and parallel processing of the conversions queue, so the next book is prepared while another is being encoded. Those two together are what make a large library practical, since the encoding is the slow part and the preparation is the fiddly part.

Smart artwork support, combined from the mp3 artwork and the cover files from the mp3 directory, which is the kind of detail that takes an afternoon to do by hand across a hundred books.

Unicode tags support, to keep tags in their own language, and advanced chapters support, including re-import of chapters from flac with cue and from m4b.

The last three are quality of life. Flexible quality management, to find the balance between size and quality. Instant playback, so you can listen to what you are going to encode before committing to it. And a simple and lightweight user interface, with the summary in the readme being that you convert a book with a few clicks.

The chapter re-import is the feature that separates this from a repackager. The source formats frequently contain the chapter information already, and preserving it is the difference between a book you can navigate and a single long track.

Three distribution channels, and a warning about two platforms

The distribution section recommends Steam first, and the reason given is worth repeating. If you are looking for a safe option to install better tested software from a trusted source and get automatic incremental updates, use Steam.

The alternatives are Chocolatey, or the latest build from the GitHub releases page, and the readme notes that no QuickTime or iTunes is required, which is a real change from the tools this kind of application used to depend on.

There is also an installation manual in the wiki, and a note that the latest features of version 4.x are shown in a short video about the chapter support.

The caveat sits at the top of the readme and applies to the two platforms that are not Windows. Mac and Linux versions are experimental and may require some skills to install and run them and may not work as you expect.

That is a fair statement rather than a dismissal. A JavaFX application with native ffmpeg dependencies, packaging conventions for the macOS bundle and the Linux desktop entry points, and a Windows-first user interface layout, is a lot of surface for a community maintainer to keep working across three platforms, and saying so plainly is better than shipping a broken build and hoping.

For an evaluation, the practical point is that the Windows build is the one to try first, whatever platform you are on.

A Maven and JavaFX project with bundled binaries

The top level listing describes a Java application built with Maven and a JavaFX interface.

There is a pom.xml as the build file, a src/ directory for the sources, an app/ directory that is probably the packaging module, a build/ directory, and a javafx/ directory that carries the interface library. There is an external/ directory for the bundled third party pieces.

The three icon files at the top are the application icons for each platform, AudioBookConverter.ico for Windows, AudioBookConverter.icns for macOS, and AudioBookConverter.png for the general case, and there is a screenshots/ directory for the images the readme and the wiki use.

Then there is the metadata. A version.txt holds the version, and there is an idea_run.txt, which is the shared run configuration for IntelliJ. The default branch is development, and the build badge points at an Azure DevOps pipeline for that branch rather than at a GitHub Actions workflow.

That last detail is a small signal about the project, because a build server that is not GitHub Actions usually means the project has been around long enough to have settled its pipeline, and the Azure setup has a build definition per branch rather than one workflow per event.

Using it is three steps, and the queue is the point

The usage section is a single sentence, and it is the design in one line. Just run the program, add files, press convert, and all other things are optional.

That simplicity is possible because the defaults are sensible, and the features behind the optional settings cover the cases where they are not. A default encode profile that produces an acceptable m4b, a default artwork lookup that finds the cover in the folder, and a default chapter scheme that uses the file order.

The queue is what makes it usable on a real library. The parallel processing entry says you can prepare the next book while encoding another, which is a different thing from parallel encoding within one book. The first is about keeping the disk and the CPU busy while you do a hundred books. The second is about a single large book being encoded across cores.

Both matter, and the second is the one that shows up in the feature list as more cores and faster results, while the first is what turns the tool from a one-off into something you leave running.

The wiki carries the longer material, with a read more link and the home page, and the video demonstrates the chapter features, which are the part of the interface that has the most surface for something to go wrong.

Editorial conclusion

Adopt AudioBookConverter when you have a library of audio files that belongs in one Apple compatible book, and you want the artwork, the chapter list and the tags handled for you. Do not adopt it as a general purpose audio converter, since its output targets are m4b, ogg opus and mp3 and the interface is built around one book at a time. Verify first which platform build you are downloading, because the Mac and Linux versions are described as experimental and may need some skill to install and run, and that the source format list covers what you actually have, since a library assembled from mixed formats is where the chapter handling has to work hardest.

Frequently asked questions

Official sources

1. Issues
2. License: GPL-2.0
3. README
4. Releases
5. yermak/AudioBookConverter on GitHub