MyMediaTools

Convert MTS to AMR — Free Online Tool

Convert MTS camcorder footage to AMR audio by extracting the AC-3 or AAC audio track and re-encoding it to Adaptive Multi-Rate format using the libopencore_amrnb codec. AMR is optimized for speech, making this conversion ideal for pulling voice narration or interview audio from AVCHD recordings for use in mobile or telephony applications.

FFmpeg Command

Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg

Free — no uploads, no signups. Your files never leave your browser.

Estimated output:

Conversion Complete!

Download

How It Works

MTS files store video and audio together in an MPEG-2 Transport Stream container, typically with H.264 video and AC-3 or AAC audio recorded at broadcast-quality bitrates. During this conversion, the video stream is discarded entirely — only the audio track is processed. That audio is then decoded from its original AC-3 or AAC encoding and re-encoded using the libopencore_amrnb codec at 12,200 bps (AMR-NB's highest quality mode). AMR-Narrowband was designed specifically for human speech frequencies (300–3400 Hz), so it applies aggressive compression that works well for voice content but will noticeably degrade music or ambient sound from your camcorder recording. The result is a compact .amr file suitable for mobile telephony systems and voice-focused applications.

What Each Flag Does

Flag What it does
ffmpeg Invokes the FFmpeg tool, which runs entirely in your browser via WebAssembly (FFmpeg.wasm) on this page — no files leave your device. The same command can be run locally on your desktop if your file exceeds 1GB.
-i input.mts Specifies the input file: an MTS file in the AVCHD format, as recorded by Sony or Panasonic camcorders. FFmpeg reads both the H.264 video stream and the AC-3 or AAC audio stream from the MPEG-2 Transport Stream container.
-c:a libopencore_amrnb Sets the audio encoder to libopencore_amrnb, which encodes AMR Narrowband — the 8 kHz, mono, speech-optimized codec used in GSM mobile telephony. This replaces the original AC-3 or AAC audio from the MTS file with a highly compressed voice-focused encoding.
-b:a 12200 Sets the AMR-NB encoding mode to 12,200 bits per second, which is the highest quality mode available in AMR Narrowband and corresponds to the GSM 12.2 kbps codec mode. This gives the best speech intelligibility the format can provide from the camcorder's original audio.
output.amr Specifies the output file as a raw AMR bitstream. The .amr extension signals that the file contains AMR-NB encoded audio without an additional container wrapper, which is the format expected by mobile telephony systems, IVR platforms, and devices with native AMR playback support.

Common Use Cases

  • Extract spoken interview or documentary narration from Sony or Panasonic camcorder MTS footage to submit to a mobile voice archiving system that requires AMR format
  • Pull voice memos or field notes recorded on a camcorder and convert them to AMR for storage in a mobile telephony or CRM platform that ingests voice data in AMR format
  • Convert AVCHD-recorded witness statements or oral history recordings to AMR for integration with legacy telecom or legal case management systems that expect narrowband voice files
  • Prepare voice-over tracks recorded on a professional camcorder for upload to an IVR (Interactive Voice Response) system that requires AMR-NB encoded audio
  • Extract and compress speech-only segments from MTS event recordings for bandwidth-efficient transfer over mobile networks where AMR's low bitrate is an operational requirement
  • Convert camcorder-recorded conference presentations or lectures to AMR for playback on older mobile handsets or embedded devices that natively support AMR but not AAC or AC-3

Frequently Asked Questions

No — AMR-NB is specifically engineered to compress human speech in the 300–3400 Hz frequency range and performs poorly on music, environmental audio, or wide-frequency sound captured by a camcorder microphone. The codec uses voice activity detection and narrow bandwidth encoding that will make music sound muffled and artificial. This conversion is only appropriate when your MTS footage contains primarily speech, such as interviews, narration, or lectures.
MTS files contain both a high-quality H.264 video stream and audio (often AC-3 at 192–384 kbps or AAC at similar rates), which together can easily consume 15–25 MB per minute. The AMR output retains only the audio, encoded at 12,200 bps — roughly 93 KB per minute. You are discarding the entire video stream and compressing audio to a fraction of its original bitrate, which accounts for the dramatic size reduction.
No. AMR is a mono, single-track format and cannot store stereo or multi-channel audio. FFmpeg will downmix your MTS audio (whether stereo, 5.1 surround, or multi-track) to a single mono channel during encoding. If your AVCHD recording has separate audio tracks — for example, a camera mic and a wireless lav mic on separate channels — only the default track will be used unless you specify a different input stream with the -map flag.
AMR-NB operates at fixed codec modes defined by the GSM standard — the numbers (4750, 5150, 5900, 6700, 7400, 7950, 10200, 12200) are bitrates in bits per second, not kilobits. The default of 12200 bps is the highest quality AMR-NB mode and is recommended for intelligible speech. To use a lower bitrate, replace the value in the command: for example, change '-b:a 12200' to '-b:a 7950' for a smaller file that still yields acceptable voice quality. Avoid modes below 5900 unless network bandwidth is critically constrained, as speech intelligibility degrades significantly.
Yes. On Linux or macOS, you can run a shell loop: 'for f in *.mts; do ffmpeg -i "$f" -c:a libopencore_amrnb -b:a 12200 "${f%.mts}.amr"; done'. On Windows Command Prompt, use: 'for %f in (*.mts) do ffmpeg -i "%f" -c:a libopencore_amrnb -b:a 12200 "%~nf.amr"'. This processes each MTS file in the folder individually, since AMR supports only a single audio track and cannot be trivially concatenated during batch encoding.
They are different codec implementations: libopencore_amrnb encodes AMR Narrowband (AMR-NB), which samples at 8 kHz and is the standard used in GSM mobile telephony; libopencore_amrwb encodes AMR Wideband (AMR-WB), which samples at 16 kHz and produces noticeably better speech quality. For most mobile telephony and legacy system compatibility, AMR-NB (the default used here) is the correct choice. If your target system explicitly supports AMR-WB and your camcorder audio contains higher-frequency speech detail worth preserving, you can substitute '-c:a libopencore_amrwb' and use a compatible bitrate like '-b:a 23850'.

Technical Notes

The libopencore_amrnb encoder imposes strict constraints that differ significantly from typical audio conversion workflows. AMR-NB requires a sample rate of exactly 8000 Hz and a single mono channel — FFmpeg will automatically resample your MTS audio (typically recorded at 48000 Hz stereo or 5.1) to meet these requirements, which means substantial audio information is discarded before encoding even begins. The 6:1 downsampling from 48 kHz to 8 kHz removes all audio content above 4 kHz, which is why the format degrades non-speech material so severely. Metadata from the MTS container — including timecode, GPS data from GPS-enabled camcorders, and chapter markers — is not carried into the AMR output format, which has no container-level metadata support. The .amr file extension typically indicates a raw AMR bitstream rather than a containerized format, so post-processing tools that expect RIFF or MP4 containers will not be able to read the output without additional conversion. If broad compatibility is your goal and AMR is not a hard requirement, extracting the MTS audio to AAC within an M4A container would preserve far more quality from the original AVCHD recording.

Related Tools

Convert MTS to ALAC Convert MTS to WTV Convert MTS to MXF Convert 3GPP to AMR Convert WebM to AMR Convert CAVS to AMR Compress MTS Extract Audio MTS to MP3