Convert DSS to AAC — Free Online Tool
Convert DSS dictation recordings to AAC audio using your browser — no upload required. This tool decodes the proprietary OKI ADPCM codec used in DSS files and re-encodes to AAC, making your voice recordings compatible with modern media players, smartphones, and transcription services.
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 low-bitrate ADPCM variant developed specifically for digital dictation hardware from Olympus, Philips, and Grundig. Because AAC uses an entirely different compression algorithm (based on psychoacoustic modeling rather than differential PCM), this conversion cannot simply remux the container — it requires a full decode-and-reencode pipeline. FFmpeg first decodes the OKI ADPCM stream to raw PCM audio, then passes that PCM through the native AAC encoder at 128 kbps to produce the output file. DSS recordings are typically sampled at low rates (often 8 kHz or 12 kHz) optimized for speech, so the AAC output will reflect the limited frequency range of the original — do not expect high-fidelity audio even at higher bitrates, since the source material itself is narrowband.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg tool. In the browser-based version of this tool, this runs as FFmpeg.wasm compiled to WebAssembly — no installation required and no files leave your device. |
-i input.dss
|
Specifies the input DSS file. FFmpeg's DSS demuxer reads the proprietary container and exposes the adpcm_ima_oki encoded audio stream for decoding. |
-c:a aac
|
Selects FFmpeg's built-in AAC encoder for the audio stream. This encoder decodes the OKI ADPCM audio from the DSS file to raw PCM, then compresses it using AAC's psychoacoustic model — a fundamentally different compression approach suited for broad device compatibility. |
-b:a 128k
|
Sets the AAC audio bitrate to 128 kilobits per second. For speech-only DSS dictation content recorded at 8–12 kHz, 128k provides more than sufficient quality; the perceptual limit for this type of narrowband audio is typically around 64–96k. |
output.aac
|
Defines the output filename and tells FFmpeg to write a raw ADTS-framed AAC file. The .aac extension is universally recognized by mobile devices, browsers, and media players that support AAC playback. |
Common Use Cases
- Uploading dictation recordings from an Olympus or Philips digital recorder to a transcription service that only accepts standard audio formats like AAC or MP3
- Archiving legacy DSS files from a corporate dictation workflow into a modern format that can be played without specialized playback software
- Importing voice memos recorded on a digital dictation device into iPhone Voice Memos, Apple Podcasts, or any iOS app that natively supports AAC
- Preparing DSS interview or field recordings for editing in a DAW or video editor that does not support the proprietary OKI ADPCM codec
- Sharing legal or medical dictation recordings with colleagues or clients who cannot install the proprietary DSS player software
- Converting a batch of archived DSS files to AAC for long-term storage on cloud platforms like Google Drive or Dropbox, where DSS files may not preview inline
Frequently Asked Questions
No — converting DSS to AAC cannot improve the audio quality beyond what was captured in the original recording. DSS files use a low-bitrate speech-optimized codec often sampled at 8 kHz or 12 kHz, which means the audio inherently lacks the upper frequency range present in music or wideband recordings. The AAC encoder will faithfully represent that limited-bandwidth audio, but setting a higher bitrate like 256k will not restore frequencies that were never recorded in the first place. The benefit of the conversion is compatibility and playback access, not quality improvement.
DSS is a proprietary format developed jointly by Olympus, Philips, and Grundig for their dictation hardware ecosystem, and it uses the OKI ADPCM audio codec which is not included in mainstream media players, operating systems, or mobile platforms. Playing DSS files typically requires manufacturer-supplied software like Olympus DSS Player or Philips SpeechExec. AAC, by contrast, is natively supported on iOS, Android, Windows, macOS, and virtually all streaming platforms, which is why converting to AAC solves the compatibility problem entirely.
Replace the value after -b:a with your desired bitrate. For example, use -b:a 64k for a smaller file that still preserves speech intelligibility, or -b:a 192k if you want more headroom. Because DSS source audio is narrowband and speech-only, values above 96k or 128k yield diminishing returns — the AAC encoder will simply have more bits to describe audio information that was never present in the DSS file. For dictation and transcription purposes, 64k is often indistinguishable from 128k.
Yes, on the command line you can use a shell loop to process multiple files. On Linux or macOS, run: for f in *.dss; do ffmpeg -i "$f" -c:a aac -b:a 128k "${f%.dss}.aac"; done. On Windows PowerShell: Get-ChildItem *.dss | ForEach-Object { ffmpeg -i $_.FullName -c:a aac -b:a 128k ($_.BaseName + '.aac') }. The browser-based tool processes one file at a time, so the local FFmpeg command is the recommended approach for bulk conversion of large DSS archives.
DSS files can embed proprietary metadata fields used by dictation management systems, such as author ID, worktype, and priority flags. These fields are part of the DSS container specification and have no standard equivalent in the AAC container format, so they will not be transferred during conversion. Standard audio tags like title or comment, if present in the DSS file as generic metadata, may or may not be mapped depending on FFmpeg's DSS demuxer implementation. If preserving dictation workflow metadata is critical, retain the original DSS files alongside the converted AAC copies.
libfdk_aac generally produces better audio quality than FFmpeg's native AAC encoder at equivalent bitrates, particularly at lower bitrates like 64k where speech intelligibility matters. However, libfdk_aac is not included in most pre-built FFmpeg binaries due to licensing restrictions, and it is not available in the browser-based FFmpeg.wasm build this tool uses. If you have a custom FFmpeg build with libfdk_aac enabled, you can substitute it by changing the command to: ffmpeg -i input.dss -c:a libfdk_aac -b:a 128k output.aac. For typical DSS speech recordings at 128k, the quality difference between the two encoders is minimal.
Technical Notes
DSS (Digital Speech Standard) files encapsulate audio encoded with the adpcm_ima_oki codec, an ADPCM variant derived from OKI Semiconductor's speech codec technology and running at very low bitrates — typically between 13 and 28 kbps. The sample rate is constrained to what the dictation hardware supports, usually 8000 Hz or 12000 Hz, which limits the audio bandwidth to approximately 4–6 kHz. When FFmpeg decodes this to PCM for re-encoding to AAC, the resulting AAC file will contain audio that sounds narrowband or telephone-quality regardless of the target bitrate, because the source information is inherently limited. The AAC container (.aac) produced by this conversion is a raw ADTS-framed AAC stream, which is broadly compatible but does not support chapters, multiple tracks, cover art, or rich metadata embedding the way an M4A container would. If you need those features — for example, to add album art or play the file in iTunes with proper tagging — consider targeting M4A instead, which uses the same AAC codec inside an MPEG-4 container. FFmpeg's native AAC encoder used here is fully free and open source; its quality is adequate for speech content but lags behind libfdk_aac for music. For DSS dictation recordings, this distinction is academic since the source material is speech at low sample rates.