Compress AU Online — Free File Size Reducer

Compress AU audio files by switching to a more efficient PCM encoding or lower-bitrate codec within the Sun AU container. This tool re-encodes your AU file using FFmpeg's pcm_s16be codec, ideal for standardizing legacy Unix audio files or reducing file size by targeting mu-law or A-law encoding.

FFmpeg Command

Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg

Free — no uploads, no signups. Your files never leave your browser.

Estimated output:

Conversion Complete!

Download

How It Works

The AU format is a simple container that stores raw PCM audio with a minimal header — it supports several codecs including 16-bit signed big-endian PCM (pcm_s16be), 8-bit PCM variants, and companded formats like mu-law (pcm_mulaw) and A-law (pcm_alaw). When compressing an AU file, FFmpeg decodes the source audio stream and re-encodes it into the target codec within a new AU container. If your source is uncompressed 16-bit PCM and you switch to mu-law or A-law (which use logarithmic companding to store audio in 8 bits), you can achieve roughly a 50% reduction in file size with only modest perceptual quality loss at the cost of a narrower dynamic range. If both source and target use the same codec, FFmpeg will still process the stream to ensure a clean, well-formed output file with a valid AU header.

What Each Flag Does

Flag What it does
ffmpeg Invokes the FFmpeg program, the open-source multimedia processing engine that handles decoding the source AU file, re-encoding the audio, and writing the output AU container with a valid header.
-i input.au Specifies the input AU file. FFmpeg reads the AU header to detect the existing audio codec (e.g., pcm_s16be, pcm_mulaw), sample rate, and channel count before beginning decoding.
-c:a pcm_s16be Sets the output audio codec to 16-bit signed big-endian PCM, which is the native default and highest-quality lossless codec supported by the AU format. This ensures the output is a standard, widely-compatible AU file that preserves full audio fidelity.
output.au Specifies the output file. The .au extension tells FFmpeg to use the Sun AU muxer, which writes the four-byte '.snd' magic number, the format header fields (encoding type, sample rate, channels), and the raw pcm_s16be audio data.

Common Use Cases

  • Reducing the size of a pcm_s16be AU file by re-encoding it to pcm_mulaw or pcm_alaw for use in telephony systems or legacy Unix applications that expect companded audio
  • Standardizing a collection of old Sun workstation AU files to a uniform 16-bit big-endian PCM format before archiving or transferring to another Unix system
  • Preparing AU audio clips for early internet streaming contexts where mu-law encoding was the expected format for .au files served from web servers
  • Converting an 8-bit unsigned PCM AU file to 16-bit signed big-endian PCM to improve dynamic range before further audio processing on a Unix pipeline
  • Cleaning up a malformed or oddly-encoded AU file by passing it through FFmpeg to produce a well-structured AU file with a correct header
  • Producing a mu-law or A-law encoded AU file for compatibility with hardware or software telephony equipment that cannot handle uncompressed 16-bit PCM audio

Frequently Asked Questions

pcm_s16be stores audio as 16-bit linear PCM in big-endian byte order, which is fully lossless but produces the largest files. pcm_mulaw and pcm_alaw are 8-bit companded formats that use logarithmic encoding — mu-law is standard in North America and Japan, while A-law is used in Europe. Both companded formats produce files roughly half the size of 16-bit PCM but introduce some quantization distortion, particularly in quiet passages where dynamic range is compressed. For telephony and voice audio, the quality difference is usually acceptable; for music or high-fidelity content, stick with pcm_s16be.
Yes, but the severity depends on the content. Mu-law and A-law encoding are designed specifically for speech telephony and perform very well for voice audio at 8-bit depth because their logarithmic curves preserve low-level signal detail. For music or audio with a wide dynamic range, however, you will notice reduced fidelity, a slightly compressed dynamic range, and some noise floor increase compared to the original 16-bit PCM. These codecs are not intended for high-quality music archiving — they are optimized for intelligible, compact voice transmission.
The AU format was developed by Sun Microsystems for use on SPARC-based Unix workstations, which use a big-endian CPU architecture where the most significant byte of a multi-byte value is stored first. The pcm_s16be codec name itself encodes this: 's16' means signed 16-bit, and 'be' means big-endian. This is in contrast to WAV, which uses little-endian PCM (pcm_s16le) because it was designed for x86 PCs. When processing AU files on modern little-endian systems, FFmpeg handles the byte-order conversion automatically.
Replace '-c:a pcm_s16be' with '-c:a pcm_mulaw' or '-c:a pcm_alaw' in the command: 'ffmpeg -i input.au -c:a pcm_mulaw output.au'. You should also ensure your audio is at 8000 Hz sample rate, as that is the standard for mu-law and A-law telephony audio — add '-ar 8000' and '-ac 1' (mono) to the command for full telephony compatibility: 'ffmpeg -i input.au -c:a pcm_mulaw -ar 8000 -ac 1 output.au'. This will produce the smallest possible AU file.
Yes, with a shell loop. On Linux or macOS, you can run: 'for f in *.au; do ffmpeg -i "$f" -c:a pcm_s16be "compressed_$f"; done'. On Windows Command Prompt, use: 'for %f in (*.au) do ffmpeg -i "%f" -c:a pcm_s16be "compressed_%f"'. This applies the same codec re-encoding to every AU file in the current directory, which is useful for batch-standardizing a large archive of legacy Unix audio files.
The AU format has an extremely minimal header — it contains only the bare essentials: a magic number, data offset, data size, encoding type, sample rate, and number of channels. There is technically an annotation field in the header that can hold a small amount of free-form text, but it is not a structured metadata system like ID3 tags in MP3 or Vorbis comments in OGG. FFmpeg may write a brief annotation string but will not embed artist, title, or album information in any standardized way, so AU is not suitable as an archival format if rich metadata preservation matters to you.

Technical Notes

The AU (Sun Audio) format is one of the oldest digital audio container formats still in circulation, defined by a four-byte '.snd' magic number followed by a short header and raw encoded audio data. Because the header includes a data offset field, the format technically supports streaming and can even represent audio of unknown length by setting the data size field to 0xFFFFFFFF — this is why it was popular for early web audio delivery. FFmpeg supports the full set of AU codecs: pcm_s16be (the default and highest quality), pcm_s8, pcm_u8, pcm_alaw, and pcm_mulaw. Re-encoding between these codecs always involves full decoding and re-encoding — there is no stream-copy shortcut available when changing codecs, even within the same container. Sample rate and channel count are preserved from the source unless you explicitly override them with -ar and -ac flags. One important limitation: the AU format does not support multiple audio tracks, chapters, subtitles, or embedded images, making it purely a single-stream audio transport format. Files processed through this tool will have a correctly formed AU header regardless of the state of the input file.

Related Tools