Convert AMR to AIF — Free Online Tool
Convert AMR audio files to AIF format by decoding the narrow-band speech-optimized AMR codec (libopencore_amrnb) into uncompressed 16-bit big-endian PCM audio (pcm_s16be), producing a lossless AIF file ready for use in Apple audio workflows. This is particularly useful when you need to bring mobile voice recordings into a Mac-based editing environment without any further audio degradation.
to
FFmpeg Command
Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg
Drop your AMR file here
or click to browse
Free — no uploads, no signups. Your files never leave your browser.
Settings
Note: Browser-based encoding uses approximate quality targets. For precise CRF compression, copy the FFmpeg command above and run it on your desktop.
Estimated output:
Conversion Complete!
DownloadHow It Works
AMR (Adaptive Multi-Rate) stores audio using a highly compressed, speech-tuned codec designed for mobile telephony — typically at bitrates as low as 4.75 kbps up to 12.2 kbps. During this conversion, FFmpeg fully decodes the AMR bitstream using the libopencore_amrnb decoder, reconstructing raw PCM audio samples from the compressed speech data. Those samples are then re-encoded as pcm_s16be — 16-bit signed PCM in big-endian byte order — and wrapped in Apple's AIF container. Because AMR is lossy, the quality ceiling of the output AIF is determined entirely by what survived the original AMR encoding: the AIF file will be lossless and uncompressed, but it cannot recover audio detail that AMR discarded when the file was first recorded. The resulting AIF file will be dramatically larger than the source AMR, since uncompressed 16-bit stereo audio at 44.1 kHz typically runs around 10 MB per minute versus a few dozen kilobytes for the equivalent AMR.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg command-line tool, which handles all decoding, sample conversion, and muxing for this AMR-to-AIF conversion entirely within your local environment (or via WebAssembly in the browser). |
-i input.amr
|
Specifies the input AMR file. FFmpeg automatically detects the AMR container and selects the libopencore_amrnb decoder to read the compressed speech bitstream encoded at one of the eight AMR-NB bitrate modes. |
-c:a pcm_s16be
|
Sets the audio codec for the output to 16-bit signed PCM in big-endian byte order, which is the standard uncompressed audio encoding for the AIF/AIFF format. This produces a lossless, uncompressed audio stream that Apple software natively reads. |
output.aif
|
Specifies the output filename with the .aif extension, which tells FFmpeg to wrap the pcm_s16be audio stream in Apple's Audio Interchange File Format container — the standard format for uncompressed audio on macOS systems. |
Common Use Cases
- Importing voice memos or call recordings made on Android or feature phones into GarageBand, Logic Pro, or Final Cut Pro, which natively read AIF files
- Transcribing mobile phone voice recordings by bringing AMR files into Mac-based transcription or audio editing software that does not support AMR input
- Archiving old AMR voice recordings from early Nokia, Sony Ericsson, or Samsung handsets into an uncompressed format for long-term preservation, since AIF is widely supported and does not depend on proprietary speech codecs
- Preparing a voice-over recorded on a mobile device for use in a professional audio post-production session on a Mac, where the editor needs an uncompressed baseline to apply EQ, noise reduction, or compression without generational lossy re-encoding
- Batch-converting corporate voicemail files exported from a PBX system in AMR format to AIF for ingestion into an audio forensics or compliance review tool running on macOS
Frequently Asked Questions
No — the AIF output will be lossless and uncompressed, but it cannot recover audio quality that AMR already discarded. AMR is a lossy format that aggressively compresses speech at bitrates between 4.75 and 12.2 kbps, and the artifacts introduced during AMR encoding (such as the characteristic narrowband 'telephone voice' at 8 kHz bandwidth for AMR-NB) are permanently baked into the audio. What AIF gives you is a stable, uncompressed container that preserves exactly what was in the AMR file — no further quality loss will occur in any subsequent editing or re-export steps.
AMR compresses speech audio down to as little as 4.75 kbps using a highly efficient lossy codec tuned for voice, while the AIF output stores raw uncompressed PCM samples. A one-minute AMR-NB file at the default 12.2 kbps bitrate is roughly 90 KB, whereas the equivalent AIF file at 16-bit PCM and 8 kHz (matching AMR-NB's sample rate) will be around 960 KB — approximately ten times larger. If the decoder upsamples to a higher rate, the file will be even larger. This size difference is expected and unavoidable when moving from a compressed to an uncompressed format.
AMR files carry very minimal metadata — typically just the file header identifying the codec mode — and do not support rich ID3-style tags or embedded contact or timestamp information in the way that formats like MP4 do. The AIF container also has limited metadata support compared to formats like FLAC or MP3. In practice, any recording date or label stored outside the AMR bitstream (such as in your phone's file system or a companion database) will not be transferred to the AIF file by this conversion.
Yes — you can replace pcm_s16be with pcm_s24be, pcm_s32be, pcm_f32be, or pcm_f64be in the command to produce AIF files with higher bit depths. For example: ffmpeg -i input.amr -c:a pcm_s24be output.aif. However, since the source AMR audio was captured at a fixed codec precision (effectively equivalent to 13-bit linear PCM for AMR-NB), using a higher bit depth will produce a larger file without recovering any additional audio detail — the extra bits will simply be empty. The default pcm_s16be is well-matched to the actual quality ceiling of AMR-NB source material.
On macOS or Linux, you can loop over all AMR files in a directory with a single shell command: for f in *.amr; do ffmpeg -i "$f" -c:a pcm_s16be "${f%.amr}.aif"; done. On Windows Command Prompt, use: for %f in (*.amr) do ffmpeg -i "%f" -c:a pcm_s16be "%~nf.aif". Each file is processed sequentially, and the output AIF files will appear alongside the originals with the same base filename. This is especially useful when exporting a batch of voicemail recordings from a PBX system or archiving a folder of legacy mobile recordings.
Yes — AIF is Apple's native uncompressed audio format and is natively supported in Logic Pro, GarageBand, Final Cut Pro, and most other macOS audio and video applications. The pcm_s16be codec used in this conversion is the standard AIF audio encoding. The one practical consideration is sample rate: AMR-NB records at 8 kHz, which is narrowband telephone quality, so while the file will import cleanly, it will sound like a phone call rather than studio audio — this is a characteristic of the original AMR source, not the conversion.
Technical Notes
AMR-NB (Adaptive Multi-Rate Narrowband), the default codec in AMR files, operates at a fixed 8 kHz sample rate with a single mono channel — it is specifically engineered for intelligible speech transmission over cellular networks, not for music or wideband audio. The libopencore_amrnb decoder used by FFmpeg implements the 3GPP AMR-NB specification and decodes all eight AMR-NB bitrate modes (4750 through 12200 bps) into linear PCM. The output pcm_s16be codec stores 16-bit signed integers in big-endian byte order, which is the byte order required by the AIF/AIFF standard (as opposed to the little-endian pcm_s16le used in WAV). Because AMR-NB is mono and 8 kHz, the output AIF will also be mono at 8 kHz by default — if you need stereo or a higher sample rate for compatibility with a specific DAW session, you can add -ac 2 -ar 44100 to the FFmpeg command, but be aware this upsampling does not create new audio information. AMR-WB (Wideband, 16 kHz) files — sometimes stored with a .awb extension or inside a .3gp container — use the libopencore_amrwb codec and will produce slightly higher-quality output since their bandwidth extends to 7 kHz. No subtitle, chapter, or multi-track metadata is present in either AMR or AIF, so there is nothing to preserve or discard in those respects.