Convert AMR to DSS — Free Online Tool
Convert AMR audio files to DSS format using the ADPCM IMA OKI codec, transforming mobile telephony voice recordings into the Digital Speech Standard used by Olympus, Philips, and Grundig dictation devices. Both formats are speech-optimized and low-bitrate, making this a natural fit for moving voice content between mobile and professional dictation workflows.
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 files use either the Narrowband (AMR-NB) or Wideband (AMR-WB) codec — both designed for compressed speech in mobile networks. DSS uses ADPCM IMA OKI encoding, a variant of Adaptive Differential Pulse Code Modulation tuned for dictation hardware from Olympus, Philips, and Grundig. FFmpeg decodes the AMR audio stream using the libopencore_amrnb or libopencore_amrwb decoder, then re-encodes the raw PCM audio into ADPCM IMA OKI and wraps it in the DSS container. Because the two codecs are entirely different, this is a full transcode — there is no stream copying. Both formats are lossy, so the conversion introduces a second generation of compression artifacts, though speech intelligibility is generally well preserved since both codecs are specifically engineered for voice.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg command-line tool, which handles all decoding, resampling, re-encoding, and container muxing for this AMR-to-DSS conversion. |
-i input.amr
|
Specifies the input AMR file. FFmpeg will automatically detect whether it is AMR Narrowband (libopencore_amrnb) or AMR Wideband (libopencore_amrwb) and select the appropriate decoder. |
-c:a adpcm_ima_oki
|
Instructs FFmpeg to encode the audio stream using the ADPCM IMA OKI codec, which is the codec used by the DSS container format for Olympus, Philips, and Grundig dictation devices. This replaces the AMR speech codec with the ADPCM-based encoding required by DSS. |
output.dss
|
Sets the output filename with the .dss extension, which tells FFmpeg to use the DSS container muxer. The combination of this container and the adpcm_ima_oki codec produces a file compatible with digital dictation device workflows. |
Common Use Cases
- Transferring voice memos recorded on a smartphone into a Philips or Olympus digital dictation workflow that only accepts DSS files
- Converting call recordings from a VoIP or mobile system that outputs AMR into DSS for transcription software that expects dictation device formats
- Archiving mobile interview recordings into DSS format for compatibility with legal or medical transcription services that use Grundig dictation equipment
- Migrating a batch of AMR voice notes into DSS so they can be played back and managed in SpeechExec or Philips dictation management software
- Preparing speech samples recorded on Android devices in AMR format for import into a DSS-based dictation review platform used by enterprise transcription teams
Frequently Asked Questions
In most cases, yes — speech intelligibility is generally well preserved. Both AMR and DSS are purpose-built for voice, not music or wideband audio, so the frequency ranges and compression strategies are compatible with spoken word content. However, because this is a lossy-to-lossy transcode, you are applying two rounds of compression, and some subtle degradation in clarity or naturalness may be noticeable, particularly if the source AMR file was already encoded at a very low bitrate like 4750 bps.
DSS with the ADPCM IMA OKI codec operates at 8000 Hz mono, which is the same sample rate used by AMR Narrowband (AMR-NB). If your source file was encoded with AMR-WB, which operates at 16000 Hz, FFmpeg will automatically downsample the audio to 8000 Hz during the transcode. This downsampling is lossless in the sense that FFmpeg handles it internally, but it does mean the wider frequency range of the WB source will be reduced to narrowband quality in the output.
No — unlike AMR, which offers selectable bitrates from 4750 bps to 12200 bps via the -b:a flag, the DSS format with the ADPCM IMA OKI codec does not expose a user-configurable quality or bitrate parameter in FFmpeg. The output bitrate is determined by the codec's fixed encoding scheme, so you cannot tune the DSS output quality the way you can with AMR. The FFmpeg command for this conversion therefore has no audio quality flag.
The command shown converts a single file, but you can batch process multiple AMR files using a shell loop. On Linux or macOS, run: for f in *.amr; do ffmpeg -i "$f" -c:a adpcm_ima_oki "${f%.amr}.dss"; done. On Windows Command Prompt, use: for %f in (*.amr) do ffmpeg -i "%f" -c:a adpcm_ima_oki "%~nf.dss". This processes every AMR file in the current directory and outputs a corresponding DSS file.
Metadata support in both AMR and DSS is extremely limited. AMR files rarely carry meaningful metadata tags, and the DSS container format has its own proprietary metadata structure that FFmpeg does not fully map to. In practice, any metadata present in the AMR source is unlikely to appear in the DSS output. If metadata preservation is important, you should document recording details externally before converting.
Compatibility depends on the specific software version. DSS is a proprietary format with several revisions (DSS and the newer DS2), and some dictation management applications like Philips SpeechExec or Olympus ODMS may expect files produced by their own hardware or companion apps, which embed device-specific metadata headers. FFmpeg-generated DSS files use standard ADPCM IMA OKI encoding and may open correctly in many players, but hardware-locked transcription platforms could reject them. Testing with your specific software is recommended.
Technical Notes
The ADPCM IMA OKI codec used in DSS is a variant of the IMA ADPCM family, adapted specifically for Olympus dictation hardware. It operates at a fixed 8000 Hz sample rate in mono, which aligns naturally with AMR Narrowband recordings but requires downsampling from AMR Wideband sources. FFmpeg's DSS muxer has limited support compared to more mainstream formats — it can write ADPCM IMA OKI audio into the DSS container but does not populate the proprietary header fields that Olympus or Philips firmware writes when a device creates a native DSS file. This means the file may lack device ID fields, author codes, or work type markers expected by some enterprise dictation management systems. There is no chapter, subtitle, or secondary audio track support in DSS, and the format does not support stereo — all output will be mono regardless of the AMR source's channel configuration. File sizes for DSS output will typically be similar to or slightly larger than low-bitrate AMR sources, since AMR's speech codec achieves very high compression efficiency compared to ADPCM-family codecs.