Initializing... drag & drop files here
Supports: WMV
sun.audio callers, and /dev/audio on classic Unix. Pick PCM A-law for European telephony interop, PCM signed 16-bit big-endian (S16BE) for full-fidelity uncompressed AU, or PCM S16LE / S24LE / S32LE for higher dynamic range. Quality Preset ranges Lowest to Highest; for bit-rate control, switch between Constant Bitrate (CBR), Variable Bitrate (VBR), or a custom bps value.WMV (Windows Media Video) is a Microsoft container/codec family introduced in 1999 whose audio track is almost always Windows Media Audio (WMA). AU is Sun Microsystems' minimal audio format — magic number 0x2e736e64 (".snd") — and was the de facto sound format on Unix workstations, NeXT machines, and the early web. Converting WMV → AU strips the video track, decodes WMA, and re-encodes the audio into a header-light file that legacy Unix toolchains, telephony pipelines, and Java audio classes can read directly.
sun.audio and javax.sound.sampled ingestion — AU is one of the small set of formats Java's built-in audio APIs decode without third-party libraries. Useful when you're embedding clips into a legacy Java applet, Swing app, or AudioInputStream consumer./dev/audio and SPARC archives — Classic Solaris and SunOS workstations exposed /dev/audio accepting µ-law AU streams. Archivists pulling audio off old SPARCstation, NeXTcube, or HP-UX disks often need WMV recordings down-converted to match..au because the format predates WAV in Unix tradition; matching their format simplifies dataset ingestion.| Property | WMV | AU |
|---|---|---|
| Type | Video container + codec family | Audio file format |
| Author | Microsoft (1999); WMV9 standardized as SMPTE 421M (VC-1) 2006 | Sun Microsystems (late 1980s) |
| Typical audio codec | Windows Media Audio (WMA v1/v2/v9) | PCM µ-law, A-law, linear PCM (8/16/24/32-bit), ADPCM |
| Header / container | ASF container (complex, object-based) | 24-byte fixed core header + payload |
| Magic number | ASF GUID 30 26 B2 75 8E 66 CF 11 A6 D9... |
0x2e736e64 (".snd") |
| Byte order | Little-endian | Big-endian (network order) |
| Telephony fit | Poor | Native (G.711 µ-law/A-law at 8 kHz) |
| Java built-in support | No (needs codec pack) | Yes (javax.sound.sampled since Java 1.1) |
| Modern OS playback | Windows native; macOS/Linux need codec packs | Most audio tools (Audacity, ffmpeg, SoX, VLC) still read it |
| Best for | Windows-era video playback | Unix audio, telephony, Java apps, legacy archives |
| Encoding | Format ID | Bit depth | Typical use | Notes |
|---|---|---|---|---|
| PCM µ-law (G.711) | 1 | 8-bit logarithmic | North American / Japanese telephony, Java applets | 8 kHz mono ≈ 64 kbit/s; default on Sun /dev/audio |
| PCM A-law (G.711) | 27 | 8-bit logarithmic | European telephony, ITU-T regions outside NA/Japan | Same bitrate as µ-law, slightly different curve |
| Linear PCM 8-bit | 2 | 8-bit signed | Low-fi voice memos | Rarely used today |
| Linear PCM 16-bit | 3 | 16-bit big-endian | CD-quality archival from AU sources | 1411 kbit/s at 44.1 kHz stereo |
| Linear PCM 24-bit | 4 | 24-bit big-endian | Studio masters | High file size; uncompressed |
| Linear PCM 32-bit | 5 | 32-bit big-endian | DSP intermediates | Lossless, maximum headroom |
You wouldn't, unless something on the receiving end specifically asks for it. The realistic reasons today are: feeding a legacy Java application that uses javax.sound.sampled without a codec dependency, sending audio into a telephony / IVR system that expects G.711 µ-law or A-law at 8 kHz, restoring archives from old Sun/NeXT/SPARC disks, or processing a research speech corpus distributed as .au. For general listening, WMV to MP3 or WMV to WAV is the better choice.
Both are 8-bit logarithmic companding schemes defined by ITU-T G.711, both produce roughly 64 kbit/s at 8 kHz mono, both are "toll-quality" speech. The difference is the compression curve: µ-law (mu-law) is used in North American and Japanese phone networks, A-law is used in European and most other countries' networks. If you're feeding a US-based Asterisk box, pick µ-law; if you're targeting European telephony, pick A-law. For non-telephony use, the two sound nearly identical.
Yes. AU is an audio-only format, so the video track of your WMV is discarded entirely. Only the audio stream (typically WMA) is decoded and re-encoded into AU. If you want to keep the video, convert to a video target instead, e.g. WMV to MP4. For other audio targets from WMV, see the general audio converter.
Yes — the Trim control on the Advanced Options panel lets you set a start time and duration in hours:minutes:seconds.milliseconds before the audio is written to AU. Only the trimmed range is encoded, which is faster than converting the whole file and trimming after. For more elaborate cuts on already-converted AU files, use the audio cutter.
8000 Hz mono with µ-law or A-law encoding — that's the G.711 telephony standard since 1972 and what Asterisk, FreeSWITCH, and most PBX/IVR systems expect. Don't upsample above 8 kHz for µ-law/A-law: it inflates the file without improving the telephony pipeline, which will downsample back to 8 kHz anyway. For non-telephony AU output, 44100 Hz is the safe general default.
AU has no built-in psychoacoustic compression — it's PCM, possibly companded to 8 bits with µ-law/A-law but never MP3-style lossy. A stereo 16-bit AU at 44.1 kHz is about 10 MB per minute. If size matters and lossy is acceptable, convert the WMV audio to MP3, AAC, or Opus instead. If you must stay in AU, drop to mono and use µ-law at 8 kHz — that yields about 480 KB per minute.
Yes. Audacity, ffmpeg, SoX, VLC, and most professional DAWs all read the AU format including µ-law, A-law, and linear PCM variants. The format hasn't changed substantially since the 1990s, so even very old tooling stays compatible. Note: some Windows-only consumer players don't recognise .au — open in Audacity or VLC if double-clicking fails on Windows.
Files are uploaded over an encrypted connection and processed on xconvert's servers, so the practical ceiling is upload size and connection speed. There's no sign-up, no per-file daily cap, and no watermark. Very long lecture recordings (multi-hour WMVs) may run faster if you split the file or use the Trim option to extract just the segment you need.
processing happens on our servers — your file isn't stored long-term on our servers, there's no account requirement, and nothing is shared with third parties. When the session ends, the working file is released. This makes WMV → AU safe for sensitive recordings such as legal depositions, internal training, or medical dictation.