AMR to OPUS Converter

Convert AMR to Opus Online

An .amr file is a voice recording stuck in a telephony codec — small, mono, and unreadable to most software that is not a phone. Opus is where that recording belongs today: an open, royalty-free codec standardised by the IETF that browsers, Android, and every modern voice app decode natively, and that is unusually good at exactly the thing AMR holds — speech at very low bit rates. This is a rescue-and-modernise conversion, not a remaster; it makes an old memo playable and tiny, but it cannot rebuild fidelity the phone never captured.

How to Convert AMR to Opus

1. Upload Your AMR File: Drag and drop the .amr onto the page or click "Add Files" to browse. Voice-recorder memos, exported voicemails, MMS clips, and call recordings all work, and several files can be queued to run with identical settings.
2. Pick a Quality Preset: Under File Compression, the Preset dropdown runs Highest, Very High (Recommended), High, Medium, Low, Very Low, Lowest. For narrowband speech a mid or low preset is plenty — the source simply has no more detail for a higher setting to capture.
3. Or Set an Exact Rate and Trim (Optional): Use Custom Bitrate, Constant Bitrate, Variable Bitrate, or Specific file size when you need to hit a number, leave Audio Channel and Audio Sample Rate on ORIGINAL to mirror the mono source, and use the Trim panel to keep only the section you want.
4. Convert and Download: Click Convert and save the .opus. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark.

AMR vs Opus at a Glance

Choosing a Bit Rate for a Narrowband Source

The mistake people make here is treating the preset as a quality dial and pushing it to the top. It is really a bit-rate budget, and the budget only needs to cover what the source contains. AMR narrowband caps out at 12.2 kbit/s of actual speech information (23.85 kbit/s if the file is a wideband .awb recording), so the ceiling is set by the recording, not the encoder.

• Speech, listening only: a Low or Medium preset. Opus is at its most efficient in exactly this range, which is why voice apps ship it.
• Speech you will transcribe or edit later: a High preset gives the encoder more headroom so nothing extra is lost before the next processing step.
• A hard size or upload limit: use Specific file size or Custom Bitrate instead of guessing with presets.
• Highest is rarely worth it here. A 128 kbit/s Opus file made from an 8 kHz phone recording is several times bigger and sounds the same as a modest one.

Where an .opus File Will Play

Frequently Asked Questions

Why convert an AMR recording to Opus instead of leaving it alone?

Because AMR is a dead end outside telephony. Very little consumer software carries an AMR decoder, the format has no metadata to record who or when, and it is patent-encumbered. Opus is the modern replacement for the same job: standardised by the IETF, royalty-free, decoded by every current browser, and tuned for exactly the low-bit-rate speech AMR holds. You end up with a small, portable file you can actually play, back up, and attach to things.

Will Opus make the recording sound better than the AMR did?

No, and it is worth being blunt about it. AMR filtered the audio to roughly the telephone band and rebuilt it from a speech model at single-digit kilobits per second; nothing downstream can restore what was discarded. Opus reproduces that narrowband speech efficiently and honestly. The genuine upside is that Opus's own speech mode is linear-prediction based — the same broad family of technique AMR uses — so it re-encodes voice at low rates with unusually little added damage.

What bit rate should I actually choose for a voice memo?

Less than instinct suggests. Since the source holds at most 12.2 kbit/s of narrowband speech information, a Low or Medium preset already carries everything there is; pushing to Highest inflates the file without adding fidelity. If you want a rule of thumb: choose the lowest preset at which you cannot hear encoder artefacts on top of the source's own telephone-quality character, and leave Audio Channel and Audio Sample Rate on ORIGINAL so nothing is resampled on the way through.

Is Opus better than MP3 for a speech recording?

For efficiency, clearly yes — Opus was designed for voice and stays intelligible at bit rates where MP3 falls apart, and it can hold sample rates as low as 8 kHz natively instead of padding them out. For compatibility, MP3 still wins on old hardware: car head units, cheap portable players, and some in-car Bluetooth stacks never added Opus. Convert to Opus for phones, browsers, and archives; use AMR to MP3 when the destination is out of your control.

What exactly is inside a .opus file?

An Opus bitstream encapsulated in an Ogg container — that combination is what the .opus extension denotes, and it is what this page writes. That matters for two practical reasons: .opus files carry Vorbis-style tags, so a recording can finally be labelled with a date and a description, and any player that understands Ogg Opus can seek and play it without extra plumbing.

Is Opus really royalty-free, including for commercial use?

Yes — that is a defining goal of the codec rather than a side effect. Opus is defined in an IETF Proposed Standard with an open specification and an open-source reference implementation, and it is deployed royalty-free in browsers, WebRTC, and messaging platforms. This is a meaningful difference from AMR, which sits under telephony patent licensing, and it is a large part of why Opus became the default voice codec on the web.

Does this handle AMR-WB (.awb) wideband recordings too?

Yes. Wideband AMR samples at 16 kHz with nine rates from 6.6 to 23.85 kbit/s, and it converts through the same path — Opus has a native 16 kHz mode, so the source rate maps across cleanly when Audio Sample Rate is left on ORIGINAL. A wideband recording will sound noticeably fuller than a narrowband one, but converting a narrowband file cannot turn it into a wideband one.

How are my files handled, and how long do you keep them?

Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark, never shared or made public. Voice recordings are usually tiny, so the whole round trip is quick; the practical limit is upload size and time rather than anything about the recording itself. If a speech-to-text tool needs uncompressed input rather than a compressed file, AMR to WAV is the better target.