Extract Audio from MPEG to AU — Free Online Tool
Extract audio from MPEG video files and convert it to Sun AU format, outputting a raw PCM 16-bit big-endian audio stream. This is particularly useful for Unix and legacy system workflows where AU's simple header structure and uncompressed PCM audio are required.
to
FFmpeg Command
Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg
Drop your MPEG 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
MPEG files typically carry MP2 (MPEG-1 Audio Layer II) audio, which is a lossy compressed format. During this conversion, the video stream is completely discarded and the MP2 audio is fully decoded and then re-encoded as PCM signed 16-bit big-endian (pcm_s16be) — the default and most common codec for AU files. This means the output is uncompressed PCM audio, so the AU file will be significantly larger than the original audio track in the MPEG file, but it will be free of any further lossy compression artifacts beyond what already existed in the MPEG source. The AU container itself uses a minimal fixed-size header storing sample rate, channel count, and encoding type, which is why it remains a staple on Unix systems and in audio processing pipelines.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg tool, which handles all media demuxing, decoding, and encoding. In the browser-based version of this tool, FFmpeg runs entirely via WebAssembly (FFmpeg.wasm) with no server upload required. |
-i input.mpeg
|
Specifies the input MPEG file. FFmpeg will demux this container, identifying the video stream (typically mpeg2video) and the audio stream (typically MP2) for separate processing. |
-vn
|
Disables video output entirely, telling FFmpeg to ignore the mpeg2video (or mpeg1video) stream from the MPEG file. Since AU is a pure audio format with no video container support, this flag is required to produce a valid output. |
-c:a pcm_s16be
|
Sets the output audio codec to PCM signed 16-bit big-endian, which is the standard and most compatible codec for the Sun AU format. The MP2 audio from the MPEG file is fully decoded and re-written as uncompressed PCM — no lossy compression is applied at the AU stage. |
output.au
|
Specifies the output filename with the .au extension, which tells FFmpeg to wrap the pcm_s16be audio stream in a Sun AU container with its characteristic minimal fixed-size header storing sample rate, channel count, and encoding type. |
Common Use Cases
- Extracting dialogue or narration audio from legacy broadcast MPEG recordings for archival on Unix-based systems that expect AU format input
- Converting MPEG training or educational video audio tracks to AU for playback or processing in older Unix workstations or scientific audio tools that natively support AU
- Supplying uncompressed AU audio extracted from MPEG source material to audio analysis or signal processing software that requires raw PCM in Sun AU format
- Stripping the MP2 audio from DVD-compatible MPEG streams and converting to AU for use in Unix shell script audio pipelines or legacy media servers
- Preparing audio assets from MPEG broadcast recordings for integration into Sun/Oracle Solaris applications or Java applets that historically relied on AU as their native audio format
- Decoding compressed MP2 audio from MPEG files into uncompressed PCM AU format to enable lossless downstream re-encoding without stacking lossy compression generations
Frequently Asked Questions
The MPEG file's MP2 audio track is already lossy-compressed, so any quality loss from that stage is permanent and already present in your source file. However, the conversion to AU with pcm_s16be does not introduce any new lossy compression — it fully decodes the MP2 stream to uncompressed PCM. The AU output is a lossless representation of whatever audio quality existed in the MPEG file, meaning no further degradation occurs during this specific conversion step.
The MPEG container stores audio as MP2, which is a lossy compressed format that significantly reduces file size. AU with pcm_s16be stores audio as raw uncompressed PCM at 16 bits per sample, which requires roughly 1.4 MB per minute per channel at 44.1 kHz — far more space than compressed MP2. For example, a 10-minute stereo MPEG with a 192 kbps MP2 track might produce an AU file of around 100 MB. This size increase is expected and is the cost of having fully uncompressed audio.
The output AU file uses pcm_s16be, which is signed 16-bit PCM with big-endian byte order. This is the default and most universally supported codec within the AU format. Most Unix and Linux audio tools, Java's AudioSystem API, and many scientific audio applications can read it natively. However, some consumer software on Windows or macOS may not support AU files directly, so if broad modern compatibility is needed, a format like WAV or FLAC might be more appropriate.
No. The Sun AU format has an extremely minimal header — it stores only technical parameters like sample rate, number of channels, encoding type, and an optional short annotation string field. It has no support for ID3 tags, Vorbis comments, or any structured metadata schema. Any title, artist, album, or other metadata embedded in the MPEG source file will be lost during conversion to AU.
You can replace '-c:a pcm_s16be' with another codec supported by the AU format. For mu-law encoding, use '-c:a pcm_mulaw', and for A-law use '-c:a pcm_alaw'. Both are 8-bit companded formats that produce smaller files but with narrower dynamic range. For 8-bit unsigned PCM use '-c:a pcm_u8', or for signed 8-bit use '-c:a pcm_s8'. The full command for mu-law, for example, would be: ffmpeg -i input.mpeg -vn -c:a pcm_mulaw output.au
Yes. On Linux or macOS you can use a shell loop: 'for f in *.mpeg; do ffmpeg -i "$f" -vn -c:a pcm_s16be "${f%.mpeg}.au"; done'. On Windows Command Prompt, use: 'for %f in (*.mpeg) do ffmpeg -i "%f" -vn -c:a pcm_s16be "%~nf.au"'. This is especially valuable for large collections or files over 1 GB, since the browser-based tool has a 1 GB per-file limit while your local FFmpeg installation has no such restriction.
Technical Notes
The MPEG container's default audio codec is MP2 (MPEG-1 Audio Layer II), a lossy format historically used in broadcast television and DVD authoring. When extracting to AU, FFmpeg fully decodes this MP2 stream in memory and writes raw PCM samples in big-endian byte order, matching the pcm_s16be codec. The AU format originated with Sun Microsystems' SunOS and remains the native audio format for Java's javax.sound.sampled API, which is why it still appears in cross-platform Java applications. One notable limitation of AU is its lack of a standardized way to store multi-channel audio beyond stereo; if the MPEG source contains more than two audio channels, behavior may vary by FFmpeg version. Additionally, AU files do not support chapters or subtitle streams, neither of which MPEG supports either, so no stream data of those types is lost. The big-endian byte order of pcm_s16be means AU files produced this way are directly compatible with SPARC and PowerPC-era Unix systems but require byte-swapping software on little-endian x86 systems if being fed to tools that do not handle endianness automatically.