Extract Audio from MTS to AMR — Free Online Tool
Extract audio from AVCHD camcorder footage (.mts files) and convert it to AMR format — a speech-optimized codec designed for mobile telephony. This tool strips the H.264 video stream and re-encodes the AC-3 or AAC audio from your MTS file into a compact AMR file using the libopencore_amrnb codec, ideal for voice-focused archiving or mobile playback.
to
FFmpeg Command
Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg
Drop your MTS 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
MTS files are AVCHD recordings from Sony or Panasonic camcorders, encapsulated in an MPEG-2 Transport Stream container with H.264 video and typically AC-3 or AAC audio. During this conversion, the video stream is completely discarded (-vn), and the audio stream is decoded from its original AC-3 or AAC encoding and then re-encoded into AMR-NB (Adaptive Multi-Rate Narrowband) using the libopencore_amrnb codec. AMR-NB operates at a fixed 8000 Hz sample rate with mono audio, so the stereo audio from your camcorder footage will be downmixed and resampled in the process. The result is a very small .amr file optimized for speech intelligibility rather than full-fidelity audio reproduction.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg tool. In the browser-based version of this tool, this runs via FFmpeg.wasm (WebAssembly) entirely client-side — no data is sent to a server. When running locally on your desktop, this calls your system-installed FFmpeg binary. |
-i input.mts
|
Specifies the input AVCHD file in MTS format, which is an MPEG-2 Transport Stream container typically holding H.264 video and AC-3 or AAC audio as recorded by Sony or Panasonic camcorders. |
-vn
|
Disables video output entirely, stripping the H.264 video stream from the MTS file so that only the audio is processed. This is what makes the tool an audio extractor rather than a video converter. |
-c:a libopencore_amrnb
|
Sets the audio encoder to libopencore_amrnb, which implements the AMR-NB (Adaptive Multi-Rate Narrowband) codec. This encoder automatically resamples the camcorder's audio to 8000 Hz mono to conform to the AMR-NB specification. |
-b:a 12200
|
Sets the AMR-NB codec mode to 12200 bps, which is the highest-quality mode available in AMR-NB and corresponds to the 12.2 kbps codec rate used in GSM-EFR telephony. This produces the most intelligible speech output from the camcorder audio. |
output.amr
|
Defines the output filename with the .amr extension, which tells FFmpeg to write a raw single-channel AMR-NB bitstream file — the standard format for AMR audio compatible with mobile devices and telephony systems. |
Common Use Cases
- Extracting voice memos or interview audio recorded on a Sony or Panasonic camcorder for use in a mobile telephony or voice messaging system that requires AMR format
- Archiving speech-only segments from AVCHD camcorder recordings — such as witness statements, oral histories, or field interviews — in a compact, mobile-compatible format
- Preparing camcorder-recorded audio for upload to telecommunications or IVR platforms that specifically accept AMR files
- Reducing storage footprint of spoken-word AVCHD recordings when video content is not needed and voice clarity at low bitrate is sufficient
- Converting camcorder-captured meeting or lecture audio into AMR for playback on older mobile handsets that natively support the format
- Generating AMR audio clips from MTS footage for integration into mobile apps or voice-driven workflows that consume AMR-NB encoded files
Frequently Asked Questions
Expect a significant reduction in audio fidelity. AMR-NB is engineered exclusively for speech intelligibility at very low bitrates — it resamples all audio to 8000 Hz mono, which is far below the 48000 Hz stereo typical of MTS camcorder audio. Music, ambient sound, or rich acoustic environments will sound notably degraded. For voice recordings like interviews or narration, the output will be intelligible but noticeably compressed. If you need higher fidelity, consider using AMR-WB (libopencore_amrwb) instead, which operates at 16000 Hz.
AMR does not support multiple audio tracks, so only the default (first) audio track from your MTS file will be converted. MTS files from some camcorders include a secondary audio channel or a separate stereo pair. If you need audio from a specific track rather than the default, you would need to modify the FFmpeg command to add '-map 0:a:1' (or the appropriate track index) before the output filename to select a different audio stream.
These numbers are the AMR-NB codec mode bitrates in bits per second, not kilobits. AMR-NB has 8 fixed codec modes ranging from 4750 bps (most compressed, lowest quality) to 12200 bps (least compressed, highest quality). The default used here — 12200 bps — is the highest available mode and gives the best speech intelligibility. Dropping to 4750 bps produces a much smaller file but audibly more robotic-sounding speech. For extracting dialogue from camcorder footage, sticking with 12200 is strongly recommended.
Replace the value after -b:a in the command with one of the valid AMR-NB mode bitrates: 4750, 5150, 5900, 6700, 7400, 7950, 10200, or 12200. For example, to use the mid-range 7400 bps mode, your command would be: ffmpeg -i input.mts -vn -c:a libopencore_amrnb -b:a 7400 output.amr. Note that AMR-NB does not support arbitrary bitrates — FFmpeg will snap to the nearest valid mode if you enter an unlisted value.
Yes, with a simple shell loop. On Linux or macOS, run: for f in *.mts; do ffmpeg -i "$f" -vn -c:a libopencore_amrnb -b:a 12200 "${f%.mts}.amr"; done. On Windows Command Prompt: for %f in (*.mts) do ffmpeg -i "%f" -vn -c:a libopencore_amrnb -b:a 12200 "%~nf.amr". This is especially practical for processing large batches of AVCHD camcorder recordings that exceed the 1GB browser limit.
No. AMR is a bare audio format with extremely limited metadata support — it does not have a metadata container equivalent to what MTS carries. Recording timestamps, GPS data, camera model information, and other AVCHD camcorder metadata embedded in the MTS file will be lost during conversion. If preserving that metadata is important, consider extracting audio to a format like M4A or Ogg that supports richer metadata fields before converting further.
Technical Notes
The libopencore_amrnb encoder enforces strict constraints that significantly shape the output: audio is always resampled to 8000 Hz and downmixed to mono, regardless of the source MTS audio characteristics. Camcorder MTS files typically contain 48000 Hz stereo AC-3 audio, so FFmpeg performs a two-step transformation — stereo-to-mono downmix and 48kHz-to-8kHz resampling — before passing the PCM data to the AMR-NB encoder. This resampling introduces an inherent frequency ceiling of ~3400 Hz in the output, which is by design for telephony use cases. The AMR format itself uses a .amr file extension for single-channel AMR-NB streams; this is distinct from 3GP or MP4 containers that can wrap AMR audio. Compatibility is broad on Android devices and older mobile handsets, but desktop media players may require codec packs to play back raw .amr files. If your end target is a modern application expecting AMR inside a mobile container, you may need an additional step to wrap the .amr stream in a 3GP container. Note also that AMR-WB (libopencore_amrwb) offers 16000 Hz bandwidth and is a meaningfully better choice when voice quality matters — it requires changing the codec flag and using its own set of valid bitrates (6600 through 23850 bps).