Initializing... drag & drop files here
Supports: 3G2, 3GP, 3GPP, ASF, AV1, AVCHD +30 more
Opus is the modern royalty-free audio codec from the Xiph.Org Foundation, standardized by the IETF as RFC 6716 in September 2012. It combines Skype's SILK speech codec with Xiph's CELT music codec into a single bitstream that adapts to whatever content you throw at it. For extracting audio from video, Opus produces noticeably smaller files than MP3 or AAC at matched perceptual quality — typically 20-30% less data for the same listening experience. Common reasons to extract video audio as Opus:
| Property | Opus | MP3 | AAC |
|---|---|---|---|
| Standardized | 2012 (RFC 6716) | 1993 (ISO/IEC 11172-3) | 1997 (ISO/IEC 13818-7) |
| Bitrate range | 6-510 kbps | 8-320 kbps | 8-529 kbps |
| Voice quality at 32 kbps | Excellent | Poor | Fair |
| Music quality at 96 kbps | Transparent for most material | Audible artifacts | Good |
| Music quality at 128 kbps | Indistinguishable from source | Good | Very good |
| Minimum algorithmic latency | 5 ms (CELT mode) | ~150-200 ms | ~20-100 ms |
| Royalty / licensing | Royalty-free, open | Patents expired April 2017 | Patents still apply |
| Mandatory in WebRTC | Yes | No | No |
| iOS Safari native playback | 18.4+ (March 2025) | All versions | All versions |
| Container | .opus (Ogg) or WebM |
.mp3 |
.m4a, .aac, .mp4 |
| Bitrate | Channels | Best for | 1-hour file size |
|---|---|---|---|
| 16-24 kbps | Mono | Voice memos, low-bandwidth speech | 7-11 MB |
| 32 kbps | Mono | Lectures, podcasts, audiobook ripping | ~14 MB |
| 48 kbps | Mono/Stereo | High-quality voice, interviews | ~22 MB |
| 64 kbps | Stereo | Casual music, mixed content | ~29 MB |
| 96 kbps | Stereo | Transparent music for most listeners | ~43 MB |
| 128 kbps | Stereo | Audiophile-grade transparency | ~58 MB |
| 192+ kbps | Stereo | Diminishing returns; archival only | 86+ MB |
Opus is roughly 20-30% more efficient than MP3 at matched perceptual quality, and the gap widens at low bitrates. A 64 kbps Opus stereo file is competitive with a 96-128 kbps MP3 for most music, and a 32 kbps mono Opus voice file beats any MP3 at the same rate. Opus is also royalty-free with no licensing footprint — useful if you're shipping audio inside an app, game, or self-hosted player.
For pure speech, 24-32 kbps mono Opus is usually all you need — that's where Opus shines hardest relative to MP3/AAC. For interviews with music intros and stings, bump to 48-64 kbps stereo. There is rarely a reason to exceed 96 kbps for voice content; the extra bits aren't audible and just bloat the file.
.opus files?Most can. Chrome, Firefox, Edge, and Opera have played Opus since 2013-2014; Android has decoded it natively since Android 5.0 (2014). iOS Safari was the last major holdout — partial support shipped in iOS 11, and full Opus support arrived in iOS Safari 18.4 (March 2025). Desktop Safari is still partial as of caniuse's current data; macOS users may prefer VLC, IINA, or foobar2000 for direct playback. If you need a format every device plays without thinking, extract to MP3 or AAC instead — see Video to MP3.
Opus is fully stereo (and supports multichannel surround up to 255 channels via channel mapping families). The "voice codec" association comes from its SILK lineage; the CELT side handles full-band music up to 48 kHz sample rate. At 96-128 kbps stereo, listening tests consistently rate Opus at or above AAC-LC and Vorbis. Set Audio Channel to Stereo if you want to preserve the source's stereo image.
Opus is internally a 48 kHz codec — the bitstream always carries 48 kHz frames, regardless of what sample rate you feed in. Lower sample rates (8/12/16/24 kHz) are advertised modes used during encoding to constrain bandwidth, not output rates. Feeding 44.1 kHz source audio works fine but causes a 44.1 → 48 kHz resample. If you have 48 kHz video audio (most modern phones and cameras record at 48 kHz), leave the sample rate as ORIGINAL or pick 48000 Hz explicitly.
Yes. Switch the Trim toggle from "Unchanged" to "Trim" and enter Start Time + Duration in seconds or HH:MM:SS.sss format. Useful for pulling a single song from a concert recording, a quote from a long interview, or a chapter from a recorded webinar. If you need a non-Opus output, see Video to MP3 or Video to AAC.
This converter re-encodes through the audio pipeline rather than stream-copying. If your source video already contains Opus audio (most WebM files from YouTube downloads do), re-encoding to Opus is essentially a transcoded copy and introduces a small additional generational loss. For lossless extraction from WebM/MKV with native Opus audio, a stream-copy tool like ffmpeg with -c:a copy is the right call. For everything else — MP4/MOV/AVI/MTS/etc. — you have to re-encode, and Opus is an excellent target.
Native iTunes and the Apple Music desktop app do not play .opus files. iPods, iPod Touches, and pre-iOS-18.4 iPhones don't play Opus either. For Apple-ecosystem distribution use AAC (.m4a) or MP3 — see Convert Video to MP3. For everything else (Android, Windows, Linux, modern browsers, Discord, game engines), Opus is the better target.
The audio extraction preserves Opus-relevant tags (title, artist, album where present) via Vorbis comments inside the Ogg container. Video chapters, subtitle tracks, and video-only metadata fields are dropped — you're keeping only the audio stream. If you need chapter-aware extraction, run the trim step once per chapter using Start Time + Duration. Once you have your .opus file you can also compress it further if you need to hit a specific upload limit.