Convert DSS to WAV — Free Online Tool
Convert DSS dictation recordings to WAV by decoding the proprietary OKI ADPCM audio codec into uncompressed 16-bit PCM — making your Olympus, Philips, or Grundig voice recordings universally playable and editable in any audio software.
to
FFmpeg Command
Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg
Drop your DSS 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
DSS files store audio using the adpcm_ima_oki codec, a variant of Adaptive Differential Pulse-Code Modulation optimized for low-bitrate speech capture on digital dictation hardware. During conversion, FFmpeg decodes this proprietary ADPCM stream sample by sample, reconstructing the underlying audio waveform, then re-encodes it as pcm_s16le — signed 16-bit little-endian PCM — inside a WAV container. This is a lossy-to-lossless-container conversion: the original DSS recording was already lossy when made, so the WAV output represents a perfect uncompressed copy of exactly what the DSS file contained. No further quality is lost. The output file will be significantly larger than the DSS source because PCM stores every audio sample explicitly rather than encoding only the differences between samples.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg binary — the open-source multimedia processing engine that handles DSS demuxing and OKI ADPCM decoding on the input side, and PCM encoding plus WAV container writing on the output side. |
-i input.dss
|
Specifies the input DSS file. FFmpeg's DSS demuxer reads the proprietary Olympus/Philips/Grundig container header and locates the adpcm_ima_oki audio stream inside it. |
-c:a pcm_s16le
|
Sets the audio codec for the output to signed 16-bit little-endian PCM — the standard uncompressed audio encoding used inside WAV files. This decodes the compressed ADPCM samples from the DSS and writes each sample as a full 16-bit value, producing a file that any audio software can read without a special codec. |
output.wav
|
Defines the output filename and tells FFmpeg to wrap the PCM stream in a WAV container. The .wav extension triggers FFmpeg's WAV muxer, which writes the standard RIFF header that Windows, macOS, and virtually all audio applications expect. |
Common Use Cases
- Importing Olympus, Philips, or Grundig dictation recordings into transcription software like Express Scribe or oTranscribe, which often require WAV or MP3 rather than DSS.
- Sending voice memos from a digital dictation recorder to a transcriptionist or legal secretary who doesn't have DSS-compatible playback software installed.
- Archiving interview recordings or field notes captured on a DSS device into an open, uncompressed format that won't depend on proprietary codec support decades from now.
- Loading dictation audio into a DAW or audio editor such as Audacity or Adobe Audition for noise reduction, normalization, or splicing before handing off to a typist.
- Feeding DSS voice recordings into speech-to-text or AI transcription APIs (such as Whisper or Google Speech-to-Text) that require standard uncompressed PCM input.
- Extracting usable audio from old DSS archives recorded years ago when the original dictation hardware is no longer functional or available.
Frequently Asked Questions
No — the WAV output will sound identical to the DSS original. DSS uses lossy OKI ADPCM compression tuned for narrow-band speech, so any quality limitations (such as reduced frequency range or compression artifacts) were introduced when the recording was made. Converting to uncompressed PCM WAV preserves exactly what was in the DSS file without adding further loss, but it cannot recover audio information that was discarded during the original recording. Think of it as making a perfect copy of an imperfect original.
DSS stores audio as OKI ADPCM, which encodes only the differences between successive audio samples at around 4 bits per sample, achieving very high compression for speech. The WAV output uses uncompressed 16-bit PCM, which stores every sample in full — typically requiring about 8 to 10 times more disk space than the DSS source. A 10-minute DSS dictation that might be 1–2 MB could expand to 10–20 MB as WAV. This is expected behavior and the trade-off for universal compatibility.
DSS files can embed dictation-specific metadata such as recording date, author ID, work type, and priority flags in their header, but the WAV container has no equivalent fields for this proprietary information. FFmpeg will extract and convert the audio stream only; dictation metadata stored in the DSS header will not be present in the output WAV file. If preserving this metadata matters, note it separately before converting, or check whether your dictation management software can export it independently.
DSS playback requires support for the adpcm_ima_oki codec, which many general-purpose media players lack. FFmpeg's built-in decoder handles this codec natively, so the browser-based tool should be able to read valid DSS files even if Windows Media Player, QuickTime, or VLC struggles with them. If the file was recorded by a non-standard or very old firmware version of a dictation device, there may be header variations that cause issues, but standard Olympus DS and Philips LFH series recordings are well supported.
Yes. DSS recordings are typically captured at 8,000 Hz or 12,000 Hz in mono — the WAV output will match those original parameters by default. If your transcription or editing software requires a specific sample rate, you can add '-ar 44100' to resample to 44.1 kHz, or '-ar 16000' for a common speech-recognition-friendly rate. To produce 24-bit output instead of 16-bit, replace '-c:a pcm_s16le' with '-c:a pcm_s24le'. The full adjusted command would look like: ffmpeg -i input.dss -ar 16000 -c:a pcm_s16le output.wav
On Linux or macOS, you can run the following in your terminal: 'for f in *.dss; do ffmpeg -i "$f" -c:a pcm_s16le "${f%.dss}.wav"; done'. On Windows Command Prompt, use: 'for %f in (*.dss) do ffmpeg -i "%f" -c:a pcm_s16le "%~nf.wav"'. This loops through every DSS file in the current directory and produces a matching WAV file with the same base name. The browser-based tool processes one file at a time, so the FFmpeg command is especially valuable when you have a large archive of dictation recordings to migrate.
Technical Notes
The DSS format was jointly developed by Olympus, Philips, and Grundig as a closed standard for professional digital dictation, and its audio codec — adpcm_ima_oki — is a proprietary variant of IMA ADPCM that uses 4-bit samples with an OKI-specific step table. DSS recordings are inherently mono and typically captured at 8 kHz (standard quality) or 12 kHz (high quality), giving them a telephone-like frequency range appropriate for voice but not music. The output WAV file using pcm_s16le will faithfully reproduce this narrow-band audio at full 16-bit resolution, though the effective frequency content remains limited to what the DSS codec captured. WAV has no meaningful file size limit for practical dictation lengths, but note that very long recordings (several hours) may require chunking for some downstream tools. FFmpeg's DSS demuxer correctly handles both the older DSS and newer DS2 variants in many cases, though DS2 files with the .ds2 extension use a different container structure and may need separate handling.