Convert VOC to DSS — Free Online Tool
Convert VOC audio files — the classic Creative Labs Sound Blaster format used in DOS-era games — to DSS (Digital Speech Standard), a compressed dictation format using the ADPCM IMA OKI codec. This tool runs entirely in your browser via FFmpeg.wasm, transcoding the raw PCM data from VOC directly to the low-bitrate, speech-optimized DSS container.
to
FFmpeg Command
Copy this command to run the same conversion locally with FFmpeg on your desktop. Download FFmpeg
Drop your VOC 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
VOC files store audio as raw, uncompressed PCM data (typically 8-bit unsigned or 16-bit signed PCM), which is the lossless, sample-exact format used by Creative Labs' Sound Blaster hardware. During conversion, FFmpeg reads the PCM audio stream from the VOC container and re-encodes it using the ADPCM IMA OKI codec to produce a DSS file. This is a full transcoding operation — the audio is decoded from PCM and then compressed with OKI ADPCM's 4-bit differential encoding, resulting in significant file size reduction. Because DSS is designed for speech dictation at narrow bandwidths, the output will be lossy and optimized for voice-frequency content, meaning any music, sound effects, or high-frequency audio present in the VOC source will be audibly degraded. The sample rate is also constrained by DSS's dictation-focused specification.
What Each Flag Does
| Flag | What it does |
|---|---|
ffmpeg
|
Invokes the FFmpeg multimedia processing tool. In the browser-based version of this tool, this runs via FFmpeg.wasm compiled to WebAssembly, executing the same conversion logic as the desktop FFmpeg binary entirely within your browser. |
-i input.voc
|
Specifies the input file — a VOC audio file in Creative Labs' Sound Blaster format. FFmpeg parses the VOC container's block headers to determine the raw PCM encoding (pcm_u8 or pcm_s16le), sample rate, and channel count before decoding. |
-c:a adpcm_ima_oki
|
Sets the audio codec to ADPCM IMA OKI, the only audio codec supported by the DSS container. This encodes the decoded PCM audio from the VOC source into 4-bit differential ADPCM compression, which is the speech-optimized, lossy encoding used by Olympus, Philips, and Grundig digital dictation devices. |
output.dss
|
Defines the output file as a DSS (Digital Speech Standard) file. FFmpeg infers the DSS container format from the .dss extension and enforces the constraints of that muxer, including the ADPCM IMA OKI codec requirement and the dictation-oriented audio parameters. |
Common Use Cases
- Archiving voice recordings or speech samples extracted from DOS-era game VOC files into a format compatible with digital dictation software from Olympus or Philips
- Repurposing legacy Sound Blaster voice-acted dialogue or narration clips from old multimedia CD-ROMs into DSS for use with professional transcription workflows
- Converting VOC-format audio memos or spoken word content stored in retro computing archives into the compact, low-bitrate DSS format for long-term storage efficiency
- Migrating speech content from early 1990s interactive fiction or edutainment software VOC assets into DSS for cataloguing in digital dictation management systems
- Testing or validating DSS decoder implementations using known PCM source material from VOC files to verify ADPCM IMA OKI encoding fidelity on speech content
Frequently Asked Questions
Yes — this is a lossy conversion. VOC stores audio as raw PCM (either 8-bit unsigned or 16-bit signed), which is lossless and sample-accurate. DSS uses ADPCM IMA OKI encoding, a 4-bit differential compression scheme designed for speech intelligibility at low bitrates, not high-fidelity reproduction. Music, sound effects, or any high-frequency audio in the VOC file will be noticeably degraded. If your VOC contains spoken voice content, the result will be more acceptable, as the DSS format was purpose-built for that use case.
This is an unusual conversion, but legitimate scenarios exist in archival and enterprise contexts. If you have speech or narration extracted from DOS games or early multimedia applications stored as VOC files, and need to ingest them into a digital dictation platform (such as Olympus Dictation Management System or Philips SpeechExec) that exclusively handles DSS files, this conversion bridges the gap. It is also useful for researchers comparing PCM source audio against ADPCM-compressed versions for codec analysis.
No. VOC files can store audio at a wide range of sample rates supported by the original Sound Blaster hardware, commonly 8000 Hz, 11025 Hz, or 22050 Hz. DSS, being designed specifically for voice dictation, operates at narrow, fixed sample rates — typically 8000 Hz. FFmpeg will handle the sample rate conversion as part of the transcoding process, but if your VOC file has a higher sample rate, it will be downsampled, which further reduces audio fidelity in the output.
Yes, you can use shell scripting to batch process files with the displayed command. On Linux or macOS, run: for f in *.voc; do ffmpeg -i "$f" -c:a adpcm_ima_oki "${f%.voc}.dss"; done. On Windows Command Prompt, use: for %f in (*.voc) do ffmpeg -i "%f" -c:a adpcm_ima_oki "%~nf.dss". The browser-based tool processes one file at a time, but the FFmpeg command is ideal for bulk operations on your local machine, especially for large archives of DOS game audio assets.
Very little metadata is preserved. VOC is a simple container format with minimal metadata support — primarily just audio encoding parameters and basic block headers. DSS is similarly constrained, being a proprietary dictation format focused on audio payload rather than embedded metadata. Fields like track title or comment tags present in the VOC file are unlikely to carry over into the DSS output. If metadata preservation is important, document any relevant information from the VOC file separately before conversion.
The core codec flag -c:a adpcm_ima_oki must remain as-is since it is the only codec supported by the DSS container format. However, you can control the output sample rate with the -ar flag — for example, adding -ar 8000 explicitly sets 8000 Hz output, which is the standard for DSS dictation files. You can also add -ac 1 to enforce mono output, which is expected for dictation use. For instance: ffmpeg -i input.voc -c:a adpcm_ima_oki -ar 8000 -ac 1 output.dss. There is no variable quality parameter for ADPCM IMA OKI, as the bit depth and compression ratio are fixed by the codec specification.
Technical Notes
The ADPCM IMA OKI codec used in DSS files is a variant of IMA ADPCM developed by OKI Semiconductor, operating at 4 bits per sample and targeting voice-band audio at 8000 Hz. The VOC container, by contrast, stores either pcm_u8 (8-bit unsigned PCM) or pcm_s16le (16-bit signed little-endian PCM) — both lossless representations. Because ADPCM IMA OKI is a fixed-ratio compression scheme, the output bitrate is determined by the sample rate rather than any configurable quality parameter, leaving no quality knob to tune. FFmpeg's DSS muxer has limited flexibility due to the proprietary nature of the format, and not all DSS-reading applications (particularly older Olympus and Philips hardware dictation recorders) may accept FFmpeg-generated DSS files, as the format has undocumented header fields that some devices validate strictly. VOC files with multiple audio blocks (a supported feature of the VOC format's chunked structure) will be decoded sequentially by FFmpeg and output as a continuous audio stream in DSS — the block boundaries are not preserved. VOC's channel and sample rate metadata embedded in its block headers is correctly parsed by FFmpeg, ensuring accurate decoding before the ADPCM re-encoding stage.