The audio file your CAT software made? We open it.

An SGMCA file is a proprietary audio format used by Stenograph's Case CATalyst software. It is an Ogg Vorbis audio container with a 96-byte proprietary header prepended at the start. DepoAudio scans the first 8 KB of the file, finds and strips that header, then passes the clean Ogg stream to FFmpeg for conversion — no quality loss.

You can't play an SGMCA file directly — media players don't recognize the 96-byte proprietary header. DepoAudio strips that header and converts the audio to WAV, MP3, FLAC, or Opus so you can open it in any player or editor. No Case CATalyst license required.

Download DepoAudio (free, Windows and macOS). Drag your SGMCA file into the window, choose WAV or MP3 from the format selector, and click Convert. Done in a few seconds per hour of audio. No command line, no FFmpeg install — it's bundled.

FTR Gold (For The Record) saves court recordings as a series of .trm files — typically one per 5-minute segment. Each .trm file contains AAC audio with a non-standard codec tag (0x4180) that most audio tools can't decode. DepoAudio handles this codec automatically and converts .trm files to standard formats.

Yes. Drop all .trm files for a session together into DepoAudio and it converts them in sequence, applying the same output settings to each. FTR uses a non-standard AAC codec tag (0x4180) that DepoAudio handles automatically.

.aes files are AES-128 encrypted and cannot be decoded without the Eclipse CAT software. To work around this, open Eclipse and export manually: File → Export Audio → WAV. Then drop the exported WAV into DepoAudio if you need to reformat or process it.

DCR files from Liberty Court Recording, High Criteria, and BIS Digital can hold up to 32 channels and need the free Liberty Court Player to open. In Liberty Court Player choose File → Export → WAV, then drop the WAV files into DepoAudio to convert, split by channel, or clean up. Everything after the export runs locally.

Support for DigitalCAT .dm audio is experimental — there is no public spec, so DepoAudio reverse-engineers it from real files. Drop a .dm file in and try a conversion; if it does not work, opening a GitHub issue with a small sample directly helps improve support. The attempt runs entirely on your machine.

Yes. DepoAudio runs natively on macOS 12 and later — both Apple Silicon and Intel Macs. The installer is a standard .dmg file. FFmpeg is bundled inside the app, so there's nothing else to install.

No. DepoAudio makes zero network calls during conversion. The entire pipeline — format detection, header stripping, FFmpeg transcoding, library storage — runs completely locally on your machine. Your client recordings never leave your computer.

Select "Split Channels" as the output mode. DepoAudio uses ffprobe to detect the channel count, then writes one output file per channel named by participant label (Reporter, Witness, Attorney 1, Attorney 2 by default — all editable). Each file also gets filed in the library under its label.

DepoAudio is free, open-source, and purpose-built for converting Stenograph SGMCA files (plus FTR, BWF, and other court recording formats). Unlike Stenograph's proprietary tools, DepoAudio requires no license, no subscription, and no account. MIT licensed — free forever.

Drag all your files into DepoAudio at once — SGMCA, FTR .trm, BWF, or standard audio files. Select your output format and settings, then click Convert. DepoAudio processes files sequentially so memory stays flat. There's no file count or size limit.

No. DepoAudio processes files sequentially so memory use stays flat regardless of batch size. FFmpeg streams audio data rather than loading it into RAM, so 8+ hour recordings convert without issues.

DepoAudio exports to WAV (lossless PCM, best for editing), MP3 at 128–320 kbps (best for sending to scopists), FLAC (lossless compressed, best for archival), Opus (voice-optimized, smallest files), and M4A/AAC (plays natively on iPhone, iPad, and Mac). You can also split multi-channel recordings into one file per participant.

BWF (Broadcast Wave Format) is a WAV file with an added "bext" metadata chunk carrying timecode and originator information. Courtroom systems like CourtSmart record in BWF. DepoAudio converts BWF to MP3, FLAC, Opus, or M4A — or re-exports plain WAV — without touching the audio quality.

Yes. The Merge tab lets you combine multiple recordings of the same event — for example, a court reporter mic plus a phone dial-in. DepoAudio automatically detects the timing offset between them and builds one clean file. You can choose "Best Quality" (picks the clearest source for each moment) or "Mix All" (blends everything together).

Click Scan and DepoAudio analyzes your audio for common issues — background noise, uneven speaker volume, clipping, narrow bandwidth. It recommends which fixes to apply and you toggle them on or off. Processing options include noise removal (Fast or Best Quality), speaker volume balancing, de-clipping, and clarity enhancement (room-echo reduction is coming in a future build). Everything runs locally on your machine.

If your computer has an Apple Neural Engine, AMD Ryzen AI (XDNA NPU), Intel AI Boost, or a discrete GPU, DepoAudio detects it automatically and uses it to speed up audio processing. If no accelerator is found, it runs on CPU — still fast, just a bit slower on very long recordings.

Every Smart Cleanup model runs on your own computer through a bundled ONNX Runtime — DepoAudio never uploads audio for AI processing. Light models ship with the app; larger and optional ones download on demand from Settings, are SHA-256 verified before use, and range from about 300 KB to 38 MB. If a hardware accelerator (Apple Neural Engine, AMD Ryzen AI, Intel AI Boost, or a GPU) is present it is used automatically; otherwise the models run on CPU.

When you select MP3 output, DepoAudio shows a bitrate choice: 128 kbps (smallest, good for email), 192 kbps (the balanced default), or 320 kbps (highest quality). Pick whichever fits per conversion — most scopists are happy with 192. For lossless audio choose WAV or FLAC instead.

To keep the installer small, only the lighter AI models ship inside the app. Larger or optional ones — speaker identification (~38 MB) and quality scoring (~300 KB) — download once, on demand, from Settings → AI Models the first time you use a feature that needs them. Each download is SHA-256 verified, and after that the model runs entirely offline.

Yes. Format conversion, cleanup with bundled models, the library, and the player all work with no internet connection. The only things that ever touch the network are optional, one-time model downloads (which you can do once on any connection) and an optional check for app updates. Your recordings are never uploaded.

Yes. DepoAudio runs natively on macOS 12 and later — Apple Silicon and Intel — and converts FTR .trm files exactly as it does on Windows. Drop the .trm segments in and export to WAV, MP3, or FLAC. Because For The Record's own software is Windows-only, DepoAudio is often the only way to open .trm files on a Mac, and it all runs locally.

Yes. Zoom H5 and H6 field recorders save standard WAV files, which DepoAudio reads directly. Convert them to MP3, FLAC, Opus, or M4A, run Smart Cleanup, or — for multi-track H5/H6 projects — choose Split Channels to keep each input as its own file. Nothing is uploaded.

A silent WAV almost always means a channel with no audio was exported on its own — for example a participant mic that was muted or never went live. Re-run the conversion with the Split Channels output mode to write every channel separately, then check each one. If the source itself plays silent in the built-in player, that track captured no sound in the original recording.

SGMCA files won't open in ordinary players because of a 96-byte proprietary header, which DepoAudio strips automatically. Drag the .sgmca file into DepoAudio and convert it to WAV or MP3 — no Case CATalyst license needed. If it still fails, the file may be truncated or only partly copied; try re-exporting it from CATalyst, or open a GitHub issue with a small sample.

Almost. Eclipse AudioSync (.aes) files are AES-128 encrypted and can only be decrypted by the Eclipse software itself, so the export starts there: File → Export Audio → WAV. From the WAV onward, DepoAudio takes over — reformat it, split channels by participant, or run Smart Cleanup, all on your own machine.

A BWF (Broadcast Wave Format) file is a WAV file — identical PCM audio — with one addition: a "bext" metadata chunk holding timecode, date, and originator details used by courtroom systems like CourtSmart. Any player that opens WAV opens BWF. DepoAudio can re-export BWF as plain WAV (dropping the extra metadata) or convert it to MP3, FLAC, Opus, or M4A.

They come from different systems and are built differently. SGMCA is Stenograph's format — Ogg Vorbis audio hidden behind a 96-byte proprietary header that blocks normal players. BWF is a standard WAV file with extra timecode metadata, used by CourtSmart and similar. SGMCA needs its header stripped before anything can read it; BWF already plays anywhere. DepoAudio converts both to standard formats.

Yes — DepoAudio is free and open source under the MIT license. There is no trial, no subscription, no account, and no per-file, watermark, or file-count limit. You can read the full source and build it yourself on GitHub. It is a personal project, not a commercial product.

Yes. Smart Audio Cleanup includes local noise removal — a Fast mode for instant results and a Best Quality mode (DeepFilterNet3) for tougher HVAC hum, paper rustle, and room noise. Click Scan and DepoAudio flags what it hears, then you toggle each fix on or off. Every model runs on your computer; no audio is uploaded.

Drop your SGMCA, FTR .trm, BWF, or other recording into DepoAudio, choose MP3, and pick a bitrate — 192 kbps is the balanced default most scopists ask for. Click Convert and the MP3 lands in your output folder and the case library. The job takes a few seconds per hour of audio, entirely on your machine.

Yes. DepoAudio processes everything locally and makes zero network calls during conversion or cleanup, so your recordings, transcripts, and case names never leave your computer — there is nothing in transit to intercept. It needs no account and no cloud upload, which keeps privileged deposition audio under your control and consistent with your duty to protect client material. Because the app is open source, those privacy claims are verifiable rather than just promised.