Initializing... drag & drop files here
Supports: AV1, MP4, MKV, WEBM
Extracting the soundtrack from an AV1 video into an Ogg file raises exactly one real decision, and it is not about quality settings — it is which codec goes inside the .ogg. The short answer: keep the default Vorbis if something downstream specifically expects Ogg Vorbis, which is still the case for a lot of game engines, internet-radio stacks and older Linux tooling. Pick Opus if the destination is modern and you simply want the best sound per kilobyte. Both are open, royalty-free codecs from the Xiph.Org Foundation, both live happily in an Ogg container, and this page writes either one.
| Codec | What it is | Compression | Pick it when | Watch out for |
|---|---|---|---|---|
| Vorbis (default here) | Xiph's original open audio codec; bitstream format frozen in May 2000 | Lossy | An importer, engine or device documents "Ogg Vorbis" support | Older and less efficient than Opus at low bitrates |
| Opus | Xiph and IETF successor codec, standardised as RFC 6716 | Lossy | You control the destination and want the best quality per kilobyte | A few legacy Ogg parsers only understand Vorbis |
| FLAC | Free Lossless Audio Codec, also from Xiph | Lossless | You want a bit-exact copy of the decoded soundtrack for archiving or editing | Much larger files; the File size (%) mode disappears because lossless size is not a dial |
| Speex | Narrowband speech codec from the early 2000s | Lossy, speech-tuned | Feeding a legacy voice system that requires it | Xiph states plainly that "the Speex codec has been obsoleted by Opus" |
.ogg target is written with Vorbis unless you change it; the Audio Codec list also offers Opus, FLAC and Speex..ogg. Files upload over an encrypted connection, convert on our servers, and are deleted automatically after a few hours. No sign-up, no watermark.| Player or browser | Ogg Vorbis |
|---|---|
| Chrome | Yes, from version 4 |
| Firefox | Yes, from version 3.5 |
| Edge | Yes, from version 17 |
| Opera | Yes, from version 10.5 |
| Safari on macOS | Full support from Safari 18.4; partial from 14.1 (caniuse) |
| Safari on iOS | Full support from iOS 18.4; partial from 17.4 |
| VLC, mpv, foobar2000 | Yes, on every desktop platform |
| Android | Yes, natively |
| Global browser coverage | About 95% of tracked traffic |
Choose Opus when you control the destination — it is the newer codec and it keeps speech intelligible at bitrates where Vorbis starts to smear. Choose Vorbis when something downstream names "Ogg Vorbis" explicitly, which still happens with game engine importers, radio automation and older Linux software that never gained an Opus decoder. There is no quality reason to prefer Vorbis; the reason is compatibility with tools that were written before Opus existed. If the destination is a browser, either works, though a standalone .opus file is the cleaner choice for pure Opus delivery — see AV1 to Opus.
On current Apple software, yes. Safari gained full Ogg Vorbis support in version 18.4 on macOS and iOS 18.4, with partial support going back to Safari 14.1 and iOS 17.4. Anything older than that will not play .ogg in the browser or in the stock media apps, though VLC for iOS and macOS handles it regardless of version. If the file has to work on an unknown or elderly Apple device, extract to AV1 to MP3 instead — MP3 is the one format nothing refuses.
Almost certainly. AV1 is a video codec, not a container: the Alliance for Open Media's 2018 bitstream specification codes pictures and never defines audio, so any sound that travels with AV1 video sits in a separate stream inside a container. In practice a file named .av1 is that container — the ones our converter produces are Matroska streams carrying the original audio alongside the video, and other tools do much the same. The only source with nothing to extract is a genuinely raw AV1 elementary bitstream, which would produce a silent .ogg.
Because FLAC is lossless, and lossless output size is dictated by the audio itself, not by a target you set. Our option set hides the percentage-based compression mode whenever the audio codec is FLAC — there is no meaningful way to promise "60% of the original" from an encoder that must reproduce every sample exactly. Quality Preset, bitrate targets and percentage sizing all come back the moment you switch to Vorbis, Opus or Speex.
Only if a legacy system demands it. Xiph's own project page states that "the Speex codec has been obsoleted by Opus" and that users are encouraged to switch, since Opus is better in every respect for speech. It remains in the dropdown because voice pipelines built in the 2000s sometimes still require Speex frames. For any new project, Opus does what Speex did and does it better at the same bitrate.
Some, because this is a re-encode rather than a copy. The audio inside an AV1 video is usually already lossy — Opus or AAC — so writing it out as Vorbis or Opus stacks a second lossy generation on the first. At the Very High preset or 128 kbps and above the difference is hard to hear on speech and most music. If you need a bit-exact copy of what the decoder produced, switch the Audio Codec to FLAC: that removes the second lossy step, though it cannot undo the first, and the file will be several times larger.
They are all Ogg containers; the extension is a hint about what is inside. .ogg is the traditional name and in practice signals Vorbis audio, .oga is the more formally correct extension for any Ogg audio stream, and .opus is reserved for Ogg files carrying Opus specifically. Most software goes by the content rather than the name, but naming a file .ogg when it holds Opus can confuse strict importers — which is why choosing Opus here is best paired with a destination you know handles it, or with a dedicated AV1 to Opus conversion.
Your upload travels over an encrypted (TLS) connection, is processed on our servers, and both the source and the finished .ogg are deleted automatically after a few hours. Files are never shared or made public, there is no account to create, and nothing is watermarked. Since the source is video and the output is audio only, the upload is by far the slow half — trimming a long recording to a Time Range shrinks the download but not the upload, so budget for the original file size when planning a big batch.