Initializing... drag & drop files here
Supports: AAC, AC3, AIF, AIFC, AIFF, AMR +14 more
A catch-all audio-to-Opus converter: bring almost any sound file — MP3, WAV, FLAC, AAC, M4A, M4B, OGG, OPUS, AIFF, AC3, AMR, AU, VOC, WEBA, or WMA — and get back an .opus file. Opus is the modern, IETF-standardized audio codec (RFC 6716, September 2012), and in blind listening tests it matches or beats MP3 on quality-per-byte at every bitrate up to transparency. The short answer: if everything that will play the file is reasonably current — a recent browser, an Android 10+ phone, a Discord/VoIP or web-streaming pipeline — convert to Opus and ship a smaller file at the same quality. If an older car stereo, a basic MP3 player, or a pre-2025 iPhone has to play it, stay on audio to MP3 instead. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically a few hours after conversion — no sign-up, no watermark, never shared or made public.
| Property | Opus | MP3 |
|---|---|---|
| Standardized | IETF RFC 6716, September 2012 | ISO/IEC MPEG-1/2 Layer III, early 1990s |
| Underlying tech | SILK (speech) + CELT (music), auto-switched | MDCT-based perceptual coding |
| Bitrate range | 6–510 kbps (CBR or VBR) | 8–320 kbps |
| Algorithmic delay | ~5–26.5 ms — usable for real-time calls | ~100+ ms — not built for live audio |
| Licensing | Royalty-free, open source | Patents expired (2017), now freely usable |
| Quality at low bitrate | Excellent — ~64 kbps rivals MP3 ~128 kbps for music | Audible artifacts below ~128 kbps |
| Sample rates / channels | 8–48 kHz, up to 255 channels | 8–48 kHz, mono/stereo |
| Browser support | Chrome 33+, Firefox 15+, Edge 14+; Safari partial, iOS Safari full from 18.4+ (~96.7% of users) | Universal — every browser and device |
| Best for | Streaming, podcasts, voice chat, web audio, storage savings | Maximum compatibility, legacy hardware, car stereos |
.opus — some still only accept MP3 or AAC.Opus chooses its internal mode automatically — SILK for speech, CELT for music, and a hybrid in between — so the right encoding bitrate depends mostly on what's in the file. The targets below follow Xiph's published recommendations; "size per hour" is the resulting file for one hour of audio.
| Target bitrate | Size per hour | Best for | Notes |
|---|---|---|---|
| 128–160 kbps | ~58–72 MB | Stereo music from a lossless master | Transparent for music per Xiph; above ~160 kbps is wasted on Opus |
| 96–128 kbps | ~43–58 MB | General music, podcasts with music beds | Xiph's recommended music range; ~128 kbps is effectively transparent |
| 64–96 kbps | ~29–43 MB | Streaming-grade music, mixed content | Clean on earbuds and phone speakers |
| 48–64 kbps | ~22–29 MB | Interviews, multi-speaker or stereo speech | Comfortable for group recordings |
| 24–32 kbps | ~11–14 MB | Single mono voice — narration, voice notes | Xiph's clean-voice range; intelligible, not for music |
| 6–16 kbps | ~3–7 MB | Narrowband/wideband speech only | RFC 6716's low floor; audible artifacts but understandable |
For music, pushing past ~160 kbps yields diminishing returns because Opus is already efficient; for a single voice, switching Audio Channel to Mono before you touch the bitrate roughly halves the size.
Better, by a measurable margin on efficiency. Opus combines two engines — SILK for speech and CELT for music — and switches between them automatically, which lets it hold quality at bitrates where MP3 starts to sound rough. In standardized blind listening tests, Opus near 64 kbps is competitive with MP3 around 128 kbps for music, and by ~128 kbps Opus is effectively transparent. It also has very low algorithmic delay (as little as ~5 ms), which is why it powers real-time calls. MP3's one lasting advantage is reach: it plays on practically every device ever made, while Opus needs reasonably modern playback support.
Some, unavoidably, because this is a lossy-to-lossy re-encode. Your MP3 or AAC source has already discarded data through its own compression, and Opus cannot rebuild detail that's already gone — it can only preserve what's left. To minimize a second round of loss, encode to Opus at a bitrate that matches or exceeds the source. The only genuinely clean, first-generation Opus encode comes from a lossless source such as WAV, FLAC, or AIFF, where the encoder works from a pristine master.
In recent browsers and on modern mobile: Chrome 33+, Firefox 15+, and Edge 14+ play Opus in the HTML audio element, and Android has played it natively since version 10. Apple was the holdout — desktop Safari has only partial support, and iOS Safari gained full Opus playback in the audio element starting with version 18.4 (early 2025). In total that's roughly 96.7% of web users today. Older car stereos, basic MP3 players, and pre-2025 iPhones often can't play it at all. If wide compatibility matters more than file size, convert audio to MP3 instead.
On the desktop, .opus is hit-or-miss in the default player: the legacy Windows Media Player and older versions of the macOS Music/QuickTime stack often won't recognize it, which is why a stray voice note downloaded from a chat app "won't open." The reliable fix is a current media player such as VLC, which plays Opus on Windows, macOS, and Linux; the built-in Media Player app on Windows 11 handles .opus too. If a colleague's software still refuses it, the safest move is to hand them an MP3 — convert Opus to MP3 for a file that opens in anything.
It depends on the content, and the bitrate guide above maps the common cases. For spoken-word audio — podcasts, audiobooks, voice notes — Opus stays clean at surprisingly low bitrates, and 24–48 kbps is often plenty. For music, aim for VBR in the 96–128 kbps range, which is typically transparent; pushing past ~160 kbps yields diminishing returns. Opus supports anything from 6 kbps up to 510 kbps, so you have room either way. If you'd rather target a file size than a bitrate, use the Specific file size option under File Compression.
It's the best choice, and usually the one those apps already use under the hood — Discord, WhatsApp voice notes, and most modern VoIP/conferencing stacks encode with Opus because its low delay and strong low-bitrate speech quality are exactly what real-time voice needs. Converting a recording to .opus before you upload keeps it small without a second wasteful transcode, and it fits comfortably under chat caps: Discord's free tier, for example, is 10 MB, and a 30-minute mono voice recording at 32 kbps lands near 7 MB. For a single speaker, 24–32 kbps mono is clean and tiny.
The codec is identical — what changes is the container. A .opus file is Opus audio wrapped in the lightweight Ogg container, which is the standard way to ship a standalone Opus track. Opus can also live inside .ogg, .webm, or .caf files, which is how it's commonly delivered for web video and streaming. This tool outputs the standalone .opus form; if you specifically need Opus audio in an OGG or WebM container for a web project, convert to those targets instead.
Yes. If every file you're converting is the same type, the dedicated pages skip straight to that pair: MP3 to Opus for the most common case, WAV to Opus for a clean lossless-to-Opus encode, or FLAC to Opus for compressed lossless masters. This generic page handles the catch-all — mixed formats, or a file whose type you're unsure of. In our testing, a 4-minute track encoded to Opus VBR at 128 kbps lands around 3.8 MB while staying perceptually close to a 320 kbps MP3 of the same audio — roughly a third of the MP3's size at comparable quality.
Your file is uploaded over an encrypted connection, processed on our servers, and deleted automatically a few hours after conversion — never shared, never made public, and not used to train any model. No account or email is required. To shrink or re-encode a mix of formats in one job instead, the broader Audio Compressor handles MP3, WAV, M4A, OGG, FLAC, and Opus together.