Initializing... drag & drop files here
Supports: CAVS
A .cavs file holds video encoded with Chinese AVS (Audio Video Standard), China's home-grown video coding standard — the kind of stream you get from Chinese set-top boxes, DVD-style players, and home-theater recordings. Almost nothing outside that ecosystem plays it, and the audio inside is locked to the video container. This tool decodes the CAVS stream on our servers and writes the sound out as a standalone MP3 — a format every phone, browser, car stereo, and music app already plays.
| Property | Value |
|---|---|
| Full name | Audio Video Standard, first generation (AVS1) |
| Developer | AVS Workgroup of China (formed June 2002, led by Gao Wen) |
| National standard | GB/T 20090.2-2006, promulgated February 2006 |
| International standard | IEEE 1857-2013 |
| Type | Lossy video compression (the .cavs file is a raw AVS1-P2 video elementary stream) |
| Coding efficiency | Comparable to H.264/AVC; roughly MPEG-2 quality at about half the bitrate |
| Broadcast profile | AVS+ (GY/T 257.1-2012), used in Chinese digital TV |
| Typical sources | Chinese set-top boxes, IPTV/DTV recordings, home-theater and DVD players |
| Decoder support | FFmpeg (cavs decoder, AVS1-P2 JiZhun profile); not natively played by Windows, macOS, or browsers |
| Best for | Broadcast and playback inside China's AVS hardware ecosystem |
| Property | Value |
|---|---|
| Full name | MPEG-1/MPEG-2 Audio Layer III |
| Standard | ISO/IEC 11172-3 (1993) and ISO/IEC 13818-3 |
| Type | Lossy audio compression (perceptual / psychoacoustic coding) |
| Common bitrates | 64-320 kbps (128 kbps speech, 192-256 kbps music, 320 kbps near-transparent) |
| Channels | Mono or stereo |
| Metadata | ID3v1 / ID3v2 tags (title, artist, album, cover art) |
| Playback | Universal — every modern phone, browser, car stereo, smart speaker, and media player |
| Best for | Distribution, mobile listening, podcasts, and sharing |
.cavs file or click "Add Files". Batch upload is supported, so you can queue several recordings at once. Files travel over an encrypted connection — the larger the file, the more upload time it takes.Because the sound is wrapped inside a Chinese AVS video stream, and AVS1 has almost no native support outside China's broadcasting hardware — Windows, macOS, and web browsers won't open a .cavs file without a special decoder such as FFmpeg. Extracting the audio to MP3 lifts the sound out of that container and puts it in a format that plays anywhere, with no codec to install.
The CAVS video container already stores its audio in a compressed (lossy) form, and MP3 is also lossy, so this is a transcode rather than a bit-perfect copy. In practice the result is hard to tell from the source if you keep a reasonable bitrate: 192-256 kbps is effectively transparent for most music, and 320 kbps is the highest MP3 setting. Only at low bitrates like 96-128 kbps will you notice softness on cymbals and reverb tails, which is usually fine for speech.
No — they're unrelated despite sharing the name. The .cavs file extension is a Chinese AVS video stream. "CAVS" is also the brand name of a karaoke software company (PlayCDG), but that software works with CD+G, MP3+G, and ZIP karaoke files, not a .cavs extension. This converter handles the Chinese AVS video case; if your file is actually a karaoke disc image, an MP3 extraction won't apply.
Match the job. For music, pick 256 or 320 kbps and leave the channel on stereo. For interviews, lectures, or other speech, 128 kbps mono cuts the file size substantially with no meaningful loss, since the human voice occupies a narrow frequency range. Leaving Audio Sample Rate on Original avoids resampling; drop it to 22 kHz only for voice-only recordings where size matters more than fidelity.
Yes. Open Advanced Options and use the Trim control to set a start point and duration, then convert — only that section is written to the MP3. This is handy for pulling a single song, segment, or clip out of a long recording instead of exporting the entire file and editing it afterward.
If you need to keep the picture, use CAVS to MP4, which rewraps the Chinese AVS stream into a widely playable MP4. If you want an uncompressed copy of the audio for editing or archiving rather than a lossy MP3, use CAVS to WAV. And if you already have an MP3 that is simply too large to share, Compress MP3 reduces its size without changing the format.
In our pipeline the .cavs file is demuxed, its AVS1 audio track is decoded, and that audio is re-encoded straight to MP3 at the bitrate you choose — there is no intermediate lossless round-trip that would add quality loss of its own. Your upload is sent over an encrypted connection, processed on our servers, and deleted automatically a few hours after conversion — no sign-up, no watermark, and your files are never shared or made public.