Convert MKV to AU — Free Online Tool
Convert MKV video files to AU audio format, extracting the audio stream and encoding it as 16-bit big-endian PCM — the native codec for Sun's classic Unix audio format. Ideal for developers and Unix system administrators who need uncompressed audio in a format compatible with legacy Sun Microsystems tools and workflows.
to
FFmpeg Command
Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg
Drop your MKV file here
or click to browse
Free — no uploads, no signups. Your files never leave your browser.
Settings
Note: Browser-based encoding uses approximate quality targets. For precise CRF compression, copy the FFmpeg command above and run it on your desktop.
Estimated output:
Conversion Complete!
DownloadHow It Works
When converting MKV to AU, the video stream is completely discarded since AU is a pure audio container with no video support. The audio stream from the MKV — which may be encoded as AAC, MP3, Opus, Vorbis, or FLAC — is decoded to raw PCM data and then re-encoded as 16-bit signed big-endian PCM (pcm_s16be). This re-encoding step is always required because AU does not support any compressed audio codecs; it stores only raw PCM waveforms. The resulting .au file contains a minimal binary header followed by the uncompressed audio samples, making it one of the simplest audio file structures in existence.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg binary, the open-source multimedia processing engine running here as WebAssembly in your browser. On your desktop, this would be the ffmpeg executable called from a terminal. |
-i input.mkv
|
Specifies the input file — an MKV (Matroska) container that may contain video, multiple audio tracks, subtitles, and chapter markers. FFmpeg will probe this file to detect all available streams before processing. |
-c:a pcm_s16be
|
Sets the audio codec to 16-bit signed big-endian PCM, which is the standard encoding for the AU format. This tells FFmpeg to decode whatever compressed audio codec is in the MKV (AAC, MP3, Opus, etc.) and re-encode the raw samples into big-endian 16-bit integers as required by AU. |
output.au
|
Specifies the output filename with the .au extension, which tells FFmpeg to write the Sun AU container format. Since AU has no video support, only the audio stream is written; all video, subtitle, and chapter streams from the MKV are automatically dropped. |
Common Use Cases
- Extracting audio from an MKV recording for playback on legacy Sun Workstations or Unix systems that natively support the .au format
- Preparing audio samples for use in Java applications, which have built-in support for AU files via the javax.sound.sampled API
- Converting MKV interview or field recordings to uncompressed AU for ingestion into older Unix-based audio editing or broadcast tools that predate modern format support
- Creating uncompressed reference audio files from MKV sources for acoustic analysis or signal processing pipelines that expect big-endian PCM in AU containers
- Stripping video from MKV game captures or tutorials to produce raw audio assets compatible with NeXTSTEP or early Solaris multimedia applications
- Archiving audio from MKV files in a format that is trivially parseable — the AU header is only 24 bytes, making it useful for low-level audio tooling and debugging
Frequently Asked Questions
It depends on what codec the audio is in inside the MKV. If the source MKV uses a lossy codec like AAC, MP3, or Opus, some quality was already lost during the original encoding, and that cannot be recovered. The AU output itself is uncompressed 16-bit PCM, so no additional lossy compression is applied — but decoding a lossy source to PCM does not restore detail that was discarded earlier. If the MKV source uses FLAC (lossless), the conversion to 16-bit pcm_s16be PCM in AU will be a true lossless representation of the original audio.
The AU format was designed in the late 1980s as a minimal audio container and has no provision for subtitles, chapters, multiple tracks, or any structured metadata beyond a basic annotation string field. All of that rich container data present in the MKV — subtitles, chapter markers, multiple audio tracks — is silently dropped during conversion. If preserving that data matters, you would need to choose an output format that supports it, such as keeping the MKV or converting to a format like MP4 or OGG.
FFmpeg will select the default or first audio track in the MKV by default. AU supports only a single audio stream, so only one track can be written. If you need a specific non-default track, you can modify the command to add '-map 0:a:1' (for the second audio track) before the output filename, like: ffmpeg -i input.mkv -map 0:a:1 -c:a pcm_s16be output.au.
By default, the output sample rate matches the source audio. You can force a specific sample rate by adding '-ar 44100' (for 44.1 kHz) or '-ar 8000' (the traditional telephone-quality rate common in AU files) to the command. The bit depth is determined by the codec choice — pcm_s16be is always 16-bit. To get 8-bit output, switch to '-c:a pcm_s8' or '-c:a pcm_u8', though this significantly reduces dynamic range.
Yes. On Linux or macOS, you can run: for f in *.mkv; do ffmpeg -i "$f" -c:a pcm_s16be "${f%.mkv}.au"; done. On Windows Command Prompt, use: for %f in (*.mkv) do ffmpeg -i "%f" -c:a pcm_s16be "%~nf.au". This processes each MKV in the current directory and produces a matching .au file.
MKV files typically store audio using compressed codecs like AAC or MP3, which can reduce audio data by 80–90% compared to raw PCM. AU stores audio as uncompressed 16-bit PCM, so a one-minute stereo audio stream at 44.1 kHz will always occupy roughly 10 MB regardless of how compact the source MKV was. This size increase is expected and is the tradeoff for having universally readable, codec-free audio data.
Technical Notes
The AU format (also known as SND or .snd on NeXTSTEP) uses a 24-byte minimum header containing a magic number (0x2e736e64), data offset, data size, encoding type, sample rate, and channel count, followed immediately by raw PCM samples. The pcm_s16be encoding means samples are stored as signed 16-bit integers in big-endian (network) byte order — a reflection of the Sun SPARC architecture's native endianness. This is a critical detail if you are reading the raw bytes programmatically on a little-endian x86/x64 machine: you will need to byte-swap the samples. The AU container has no support for subtitles, chapters, metadata tags beyond a simple annotation field, or multiple audio tracks — all of which MKV supports natively. Any such content in the source MKV is discarded without warning. There is also no standard way to store floating-point or high-resolution (24-bit, 32-bit) PCM in an AU file that is broadly supported, making it unsuitable as an archival format for high-resolution audio. For modern archival use, FLAC or WAV would be more appropriate; AU's value is specifically in its Unix legacy compatibility and its trivially simple structure.