Convert MPEG to AMR — Free Online Tool
Convert MPEG video files to AMR audio, extracting and re-encoding the audio track using the Adaptive Multi-Rate Narrowband codec (libopencore_amrnb) — the same speech-optimized format used in mobile telephony. This is ideal when you need to isolate spoken dialogue or voice content from MPEG footage for use in voice-driven mobile or telephony applications.
to
FFmpeg Command
Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg
Drop your MPEG 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
MPEG files typically carry MP2 or MP3 audio alongside MPEG-1 or MPEG-2 video streams. During this conversion, the video stream is discarded entirely — AMR is a pure audio format with no video container support. The audio track is decoded from its original MP2/MP3 encoding and then re-encoded using the libopencore_amrnb codec into the AMR Narrowband format. AMR-NB operates at fixed bitrates between 4.75 kbps and 12.2 kbps, far below typical broadcast audio bitrates, and its codec is tuned specifically for human speech frequencies (300 Hz–3400 Hz). This means music, sound effects, or ambient audio from the MPEG source will lose significant fidelity, while clean speech content will remain intelligible.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg tool. In the browser, this runs via FFmpeg.wasm (WebAssembly), so no installation is needed and no files leave your device. On your desktop, this requires a local FFmpeg installation. |
-i input.mpeg
|
Specifies the input MPEG file, which may contain MPEG-1 or MPEG-2 video alongside MP2, MP3, or AAC audio. FFmpeg reads both streams but only the audio will be used in the output. |
-c:a libopencore_amrnb
|
Sets the audio codec to libopencore_amrnb, which encodes the decoded MPEG audio into AMR Narrowband format — the standard speech codec used in GSM mobile telephony, operating at 8000 Hz mono. |
-b:a 12200
|
Sets the audio bitrate to 12.2 kbps, the highest quality mode available in AMR-NB. This corresponds to the AMR mode 7 (MR122) and gives the most intelligible speech output. AMR-NB only accepts specific fixed bitrate values; 12200 is the default and recommended setting. |
output.amr
|
Defines the output filename with the .amr extension, which signals FFmpeg to write a raw AMR-NB bitstream container. The video stream from the MPEG source is automatically dropped because the AMR format supports audio only. |
Common Use Cases
- Extracting spoken commentary or narration from legacy MPEG broadcast recordings for use in IVR (Interactive Voice Response) telephony systems that require AMR-formatted audio prompts
- Converting MPEG news archive footage audio tracks to AMR for storage and playback on older mobile handsets or feature phones that natively support AMR but not MP2 or MPEG video
- Pulling interview dialogue from MPEG video files to create AMR voice clips compatible with mobile messaging or voice note applications
- Preparing speech content recorded in MPEG format for submission to telecom or voice AI platforms that require low-bitrate AMR-NB input files
- Archiving the spoken audio layer of DVD-compatible MPEG recordings in a compact AMR format to minimize storage when video is no longer needed
- Converting MPEG-captured lecture or conference recordings into AMR files for distribution via SMS-based or WAP-era mobile content delivery systems
Frequently Asked Questions
No — AMR-NB is specifically engineered for human speech and performs poorly on music, environmental audio, or sound effects. The codec uses a narrowband frequency range of 300 Hz to 3400 Hz, which discards most of the richness in non-speech audio. If your MPEG file contains music or mixed audio, you will notice heavy artifacts and tonal distortion. AMR is the right choice only when the audio content is primarily spoken voice.
Two things happen simultaneously: the video stream (which accounts for the vast majority of an MPEG file's size) is completely dropped, and the remaining audio is encoded at a maximum of 12.2 kbps — compared to the 192 kbps or higher bitrates typical in MPEG audio tracks. An MPEG file that is hundreds of megabytes may produce an AMR file measured in kilobytes, purely because AMR-NB operates at some of the lowest bitrates of any standardized audio codec.
AMR-NB (Narrowband) tops out at 12.2 kbps and covers the 300–3400 Hz speech range, which is the classic telephone audio quality. AMR-WB (Wideband), also called G.722.2, extends up to 23.85 kbps and covers 50–7000 Hz, producing noticeably clearer speech. This tool uses AMR-NB by default via libopencore_amrnb, matching the most widely deployed mobile telephony standard. If you need wideband quality, you can modify the FFmpeg command to use libopencore_amrwb and an .awb output extension.
Regardless of whether the source MPEG uses MP2, MP3, or AAC audio, the output quality is ultimately constrained by the AMR-NB codec and its maximum 12.2 kbps bitrate — not by the source codec. FFmpeg fully decodes the source audio to raw PCM before re-encoding to AMR, so the intermediate codec type has minimal impact. A higher-quality source audio track may marginally reduce accumulated decoding artifacts, but the AMR encoding step is the dominant quality bottleneck.
Replace the value after -b:a with one of the AMR-NB fixed bitrate modes: 4750, 5150, 5900, 6700, 7400, 7950, 10200, or 12200 (in bits per second, not kbps notation). For example, use -b:a 7950 for a mid-range speech quality. Note that AMR-NB does not support arbitrary bitrates — FFmpeg will snap to the nearest valid mode. The default of 12200 (12.2 kbps) gives the highest AMR-NB quality and is recommended unless storage is extremely constrained.
Yes. On Linux or macOS you can use a shell loop: for f in *.mpeg; do ffmpeg -i "$f" -c:a libopencore_amrnb -b:a 12200 "${f%.mpeg}.amr"; done. On Windows Command Prompt, use: for %f in (*.mpeg) do ffmpeg -i "%f" -c:a libopencore_amrnb -b:a 12200 "%~nf.amr". The browser-based tool processes one file at a time, so the FFmpeg command approach is especially practical when converting large batches of MPEG archive files.
Technical Notes
AMR-NB enforces mono audio at an 8000 Hz sample rate — if your MPEG source contains stereo audio (common in broadcast or DVD-compatible MPEG files), FFmpeg will automatically downmix to mono before encoding. This is a hard codec constraint, not a tool limitation. The .amr file container stores only a single audio track and carries no support for chapters, subtitles, or metadata tags beyond basic header information, so any embedded metadata in the MPEG source is not preserved. Because both MPEG audio and AMR use lossy compression, this conversion involves a decode-then-re-encode cycle that introduces generation loss — the quality degradation is most noticeable at AMR's lower bitrate modes. The libopencore_amrnb library implements the 3GPP AMR-NB standard and produces files compatible with virtually all devices and platforms that support AMR, including legacy Nokia and Sony Ericsson handsets, Android, iOS voice systems, and telecom infrastructure. One important limitation: if your MPEG file has multiple audio tracks, only the default (first) audio track will be extracted and converted.