Convert MKV to AMR — Free Online Tool
Convert MKV video files to AMR audio, extracting and re-encoding the audio track using the libopencore_amrnb codec optimized for speech. AMR (Adaptive Multi-Rate) is a narrow-band audio format designed for voice telephony, making this conversion ideal when you need to strip spoken-word content from a video for use in mobile or telecom contexts.
to
FFmpeg Command
Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg
Drop your MKV 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
During this conversion, the video, subtitle, and chapter streams from the MKV container are entirely discarded — AMR is an audio-only format with no support for video or metadata containers. The audio track is decoded from its original MKV codec (typically AAC or another lossy format) and then re-encoded using the libopencore_amrnb encoder at 12,200 bps. Because AMR Narrowband operates at an 8 kHz sample rate and is tuned specifically for the frequency range of human speech, any audio content outside that range — such as music, wide-frequency sound effects, or stereo information — will be significantly degraded. This is a lossy-to-lossy transcode, meaning there is a generation loss compared to the original audio.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg media processing tool. In this browser-based tool, FFmpeg runs entirely via WebAssembly (FFmpeg.wasm) inside your browser — no data is sent to a server. |
-i input.mkv
|
Specifies the input file in Matroska (MKV) format. FFmpeg will demux all streams — video, audio, subtitles, and chapters — from the MKV container, though only the audio stream will be used in this conversion. |
-c:a libopencore_amrnb
|
Sets the audio encoder to libopencore_amrnb, the open-source implementation of the AMR Narrowband codec. This encoder forces 8 kHz mono output regardless of the source audio's sample rate or channel count, as required by the AMR-NB specification. |
-b:a 12200
|
Sets the AMR Narrowband bitrate to 12,200 bps, which is the highest available mode for AMR-NB and produces the best speech quality the format can deliver. AMR-NB bitrates must be one of eight fixed values defined by the 3GPP standard; arbitrary bitrates are not supported. |
output.amr
|
Defines the output file with the .amr extension, which tells FFmpeg to write a raw AMR bitstream. The AMR format carries no video, subtitle, chapter, or rich metadata — only the encoded speech audio stream. |
Common Use Cases
- Extracting spoken dialogue from an MKV lecture recording to upload as a voice memo or voice message on a mobile device that natively plays AMR files
- Converting a narrated MKV tutorial or walkthrough into AMR format for storage and playback on older feature phones or embedded telecom systems that only support AMR
- Isolating a speech track from an MKV interview recording to submit to a voice transcription or speech recognition pipeline that accepts AMR input
- Reducing file size of spoken-word MKV content (e.g., voiceover recordings, dictation) to the absolute minimum for transmission over bandwidth-constrained mobile networks
- Extracting a phone call recording that was captured and wrapped in an MKV container back into native AMR format for compatibility with telephony archiving tools
- Preparing voice-only audio from an MKV court or deposition recording for use in legal case management systems that expect AMR-formatted audio files
Frequently Asked Questions
No — AMR Narrowband is specifically engineered for the 300–3400 Hz frequency range of human speech and samples audio at only 8 kHz. Music, sound effects, or any wideband audio content will sound heavily muffled, hollow, and distorted after conversion. This conversion should only be used when the MKV audio track is predominantly speech or voice narration.
AMR is a raw audio-only format — it has no container structure capable of holding video streams, subtitle tracks, chapter markers, or metadata. Everything except the primary audio track is automatically dropped during conversion. If you need to preserve subtitles or chapters, you would need to choose a container format like MKV, MP4, or WebM as your output instead.
AMR Narrowband (NB) operates at an 8 kHz sample rate and bitrates from 4,750 to 12,200 bps, designed for standard GSM telephony. AMR Wideband (WB) operates at 16 kHz and supports higher bitrates, producing noticeably better speech quality. This tool uses libopencore_amrnb by default, which produces the most universally compatible AMR files for older mobile devices and telecom systems. If your target system supports AMR-WB, you would need to modify the FFmpeg command to use libopencore_amrwb instead.
By default, FFmpeg selects the first audio track (stream index 0:a:0) in the MKV file for conversion. If your MKV has multiple audio tracks — for example, a director's commentary alongside the main dialogue — and you want to extract a specific one, you would need to modify the FFmpeg command by adding '-map 0:a:1' (or the appropriate stream index) before the output filename.
The '-b:a 12200' flag controls the bitrate. AMR Narrowband supports only eight fixed bitrates: 4750, 5150, 5900, 6700, 7400, 7950, 10200, and 12200 bps — you must use one of these exact values, as AMR does not support arbitrary bitrates. The default of 12200 bps gives the best speech quality for AMR-NB. To reduce file size at the cost of intelligibility, you could use a lower value such as '-b:a 7950', but speech quality degrades noticeably below that.
Significantly smaller. A typical MKV with AAC audio at 128 kbps will have its audio track reduced to just 12.2 kbps in AMR-NB — roughly a 10x reduction in audio data. Combined with the complete removal of the video stream (which is usually the largest component of an MKV), the resulting AMR file will often be 50–200x smaller than the source MKV. For example, a 500 MB MKV with 60 minutes of audio would produce an AMR file of roughly 5–6 MB.
Technical Notes
AMR Narrowband enforces strict constraints that make it fundamentally different from general-purpose audio formats. The encoder forces a mono channel (stereo input is automatically downmixed) and resamples all audio to exactly 8,000 Hz before encoding. This means if your MKV's audio track is stereo at 48 kHz — the most common configuration — FFmpeg will perform automatic sample rate conversion and channel downmixing as part of the transcode pipeline. There is no way to preserve stereo or high-fidelity audio in an AMR-NB file by design. Additionally, because AMR uses lossy compression and your MKV likely already contains a lossy-encoded audio stream (AAC or MP3), this conversion is a second-generation lossy encode. For best results, if you have access to a lossless or high-bitrate version of the audio, prefer converting from that source. The libopencore_amrnb library used by FFmpeg.wasm is an open-source implementation of the 3GPP AMR standard and produces files compatible with most mobile platforms and telecom systems. Note that AMR files use the .amr extension and lack a rich container header, so metadata such as track title, artist, or album from the original MKV is not preserved in the output.