AU to WAV Converter

How to Convert AU to WAV Online

1. Upload Your AU File: Drag and drop or click "Add Files" to select.au or.snd files. Sun/NeXT workstation recordings, classic Java sound assets (Java 1.0-era applets shipped.au), scientific signal captures, and SPARCstation desktop sounds all work. Batch is supported — convert an entire archive folder in one pass.
2. Pick the WAV PCM Codec: Default is PCM_S16LE (16-bit little-endian, the universal CD-standard WAV variant). Choose PCM_S24LE for 24-bit studio depth, PCM_S32LE for 32-bit archival headroom, PCM_S16BE for big-endian PCM, or PCM_MULAW / PCM_ALAW if you need to preserve telephony-style 8-bit log encoding inside a WAV container.
3. Set Sample Rate, Channels, and Trim (Optional): Choose 8 kHz (the AU default for legacy μ-law speech), 11.025 / 22.05 / 44.1 kHz (CD-family), or 48 / 96 kHz (video / studio). Pick mono or stereo via AUDIO_CHANNEL. Optionally trim using start time + duration in HH:MM:SS.sss format to extract a single sample from a longer.au capture.
4. Convert and Download: Click Convert. Files convert on our servers and download individually or as a ZIP — no sign-up, no watermark.

Why Convert AU to WAV?

AU (also known as the Sun/NeXT audio format, with extension.au or.snd) is a legacy audio container introduced by Sun Microsystems in the late 1980s for Unix workstations and adopted as the native sound format of early Java (the sun.audio API only played.au). It's a tiny header followed by raw audio samples — usually 8-bit μ-law at 8 kHz, sometimes 16-bit linear PCM. WAV is Microsoft's RIFF-based PCM container, the universal interchange format for uncompressed audio. Common reasons to convert AU → WAV:

• Legacy Unix workstation archives — SPARCstation, NeXTcube, and SGI Indy systems recorded everything as.au. Migrating decades-old audio collections (corporate voicemail, university recordings, research data) to WAV makes them playable in Windows Media Player, QuickTime, and every modern DAW without specialty codecs.
• Scientific recordings and bioacoustics datasets — Older ornithology corpora, seismic monitoring outputs, and lab signal captures published in the 1990s-2000s frequently used.au at 8 kHz μ-law. WAV is what MATLAB, Python (scipy.io.wavfile), R, and Praat read natively.
• Java sound assets — Original Java 1.0/1.1 applets only played AU via sun.audio.AudioPlayer. Modernizing those assets for use in JavaFX, Web Audio, Unity, or current Android requires WAV (or MP3/OGG).
• DAW import for restoration — Audacity, Pro Tools, Reaper, and Logic Pro accept WAV uniformly. AU support varies — Logic Pro X dropped it, and many VST loaders refuse.au outright. Convert first, edit second.
• CD burning and broadcast playout — Red Book audio CD masters require 44.1 kHz / 16-bit PCM WAV. Broadcast automation systems (RCS Zetta, ENCO, Wide Orbit) accept WAV but reject AU.
• Email, Discord, and modern messaging — Operating systems and chat clients show.au as an unknown attachment. WAV plays inline in Slack, Discord, iOS Files, Android, and macOS Preview.

AU vs WAV — Format Comparison

WAV PCM Codec Choice

Frequently Asked Questions

Is converting AU to WAV lossless?

It depends on what's inside the AU. If the AU contains 16-bit linear PCM and you output 16-bit PCM WAV at the same sample rate, the conversion is lossless — every sample value is preserved. If the AU contains 8-bit μ-law (the most common case for legacy Sun files), the audio is decoded from log-companded 8-bit into linear PCM, which is mathematically faithful but can't restore dynamic range that μ-law's compander already discarded at recording time. To keep the original encoding style, choose PCM_MULAW as the WAV codec — the audio is still decoded and re-encoded on our servers (there is no byte-for-byte copy mode), but the output stays 8-bit μ-law rather than being expanded to linear PCM.

Why is my AU file only 8 kHz mono?

Because Sun's classic.au default — and Java's sun.audio API default — was 8 kHz, 8-bit μ-law, mono. That choice came from the telephony origins of μ-law (G.711) and the storage constraints of late-1980s workstations. Converting to WAV doesn't add fidelity that wasn't there: the upper limit is around 4 kHz of audio bandwidth (the Nyquist of an 8 kHz sample rate). For these files keep 8 kHz mono in the WAV output unless a downstream tool requires 44.1/48 kHz, in which case upsampling is fine but doesn't add detail.

Should I keep the AU file or delete it after converting?

Keep it. WAV is the universal modern format, but the original AU is a small archival artifact — usually a few hundred kilobytes — and historians, museum archivists, and software preservation projects often want the bit-original. A safe pattern: convert to WAV for everyday use, archive the.au alongside a small text note documenting the source workstation and recording context.

What about the.snd extension — is it the same thing?

Mostly yes. .snd was used on NeXT systems for the same Sun/NeXT audio container, and on classic Mac OS for an unrelated System 7 sound resource format. XConvert's AU pipeline handles the Sun/NeXT .snd (the common case for Unix archives). If your .snd is a Mac System 7 resource, you'll need a Mac-specific extractor first — those are rare and pre-OS-X.

Can I convert AU to WAV preserving μ-law to load into a telephony system?

Yes — pick PCM_MULAW or PCM_ALAW as the WAV codec. The WAV container then holds 8-bit log-companded samples that telephony platforms (Asterisk, FreeSWITCH, Twilio Programmable Voice prompts) accept directly. The converter still decodes and re-encodes internally, but delivering μ-law/a-law output keeps the file in the telephony domain end-to-end instead of committing it to 16-bit linear PCM that the phone platform would only re-encode again.

Do Java 1.0.au applet sounds still play on modern systems?

Not in browsers — Java applets are gone. But javax.sound.sampled in modern Java still reads.au, so the audio data isn't lost; it just needs a container migration. Converting to WAV (or WAV to MP3 for size reduction) makes the assets usable in modern web pages via <audio>, in Android SoundPool, and in game engines.

Why does the WAV file sometimes sound slightly different from playing the AU directly?

Two reasons. First, some media players (especially older Windows ones) auto-resample on playback to match the system mixer rate (typically 48 kHz), introducing a small filter coloration that the WAV won't have if you keep the source rate. Second, μ-law decoding to linear PCM is mathematically defined but the choice of dither when truncating back to 16-bit PCM differs by tool. For critical fidelity, choose PCM_MULAW output so the stored samples stay 8-bit μ-law and no 16-bit truncation is baked into the file.

Can I batch convert a whole directory of legacy.au archive files?

Yes — drop the entire folder. Each file converts in parallel on our servers and downloads individually or as a single ZIP. Settings apply uniformly, which is ideal when migrating a corpus that was captured with the same workstation and recording chain. For huge collections, consider also AU to MP3 if you only need listening copies and want 10× smaller files.