In memory
Audio is captured while the hotkey is active, transcribed locally, then discarded.
Privacy
Privacy model
Presspeech is local-first in the strict sense: no analytics SDK, no accounts, no crash reporter, no transcript sync, and no cloud transcription endpoint.
What happens to the audio
Transcript handling
Paste at cursor
The finished text normally goes through the platform clipboard and a synthetic paste shortcut. On macOS, Presspeech falls back to direct Unicode key events if clipboard paste setup fails.
Clipboard persistence
Each recording is bound to the window focused when it began. If focus changes or Presspeech cannot verify the same destination, it leaves the transcript on the clipboard for deliberate manual paste instead of risking delivery to the wrong window. macOS leaves the transcript on the pasteboard by default. The opt-in Restore clipboard after paste setting takes a guarded, best-effort snapshot and restores it only when complete and still current, so it does not overwrite a newer copy. If a slow app receives old clipboard content, turn that setting off. Windows leaves the transcript on the clipboard until something else is copied.
Operating-system clipboard services
“No transcript sync” describes Presspeech itself, not services that monitor the general clipboard. With Handoff enabled, macOS Universal Clipboard can make copied content available on nearby Apple devices. Windows clipboard history can retain copied items and its cross-device setting can sync them through the signed-in account. Third-party clipboard managers may also retain entries. Review System Settings → General → AirDrop & Handoff on macOS or Settings → System → Clipboard on Windows before sensitive dictation.
No transcript logs
Logs include timing and length metadata only. Transcript content is not written to disk.
Four call types, fully listed
Network calls
This is the complete network surface of Presspeech itself and its install/update path. Operating-system clipboard sync and third-party clipboard managers are separate services described above.
| When | Destination | Purpose | User control |
|---|---|---|---|
| Model download | huggingface.co |
macOS downloads the FluidAudio Parakeet TDT v3 CoreML model (about 500-600 MB) on first launch, after an integrity failure, or after a user-triggered cache reset. Windows downloads its selected local model; fresh systems use Parakeet (about 2.5 GB) when CUDA is usable or Whisper base.en (about 141 MiB) on CPU. No dictation audio is sent. If HF_TOKEN, HUGGING_FACE_HUB_TOKEN, or HUGGINGFACEHUB_API_TOKEN is set, FluidAudio may attach it to macOS model requests; most users have none set. |
Required for each platform's local transcription. |
| Periodic update check | api.github.com/repos/rcourtman/presspeech/releases/latest (macOS); api.github.com/repos/rcourtman/presspeech/releases?per_page=100 (Windows) |
macOS checks every six hours; Windows checks at most daily. Both only notify and use fixed platform User-Agents (presspeech-update-check and presspeech-windows-update-check) with no version, device, settings, transcript, or usage identifiers. |
Disable update notifications from Settings. |
| User-triggered feedback form | github.com/rcourtman/presspeech/issues/new |
Report a Problem and Suggest an Improvement open fixed public GitHub forms in the default browser. Presspeech adds no version, device, settings, transcript, diagnostics, or usage data to the URL; the browser and GitHub handle their normal request and account metadata. | Only runs after the user chooses one of the feedback actions. |
| User-triggered install/update | github.com/rcourtman/presspeech/releases/download/..., formulae.brew.sh, github.com/rcourtman/homebrew-presspeech, Homebrew's own update endpoints (github.com / api.github.com) |
macOS uses the signed, notarised archive directly or through Homebrew; each immutable GitHub release also publishes a SHA-256 sidecar. Windows downloads an installer only after user approval, then verifies the published size and SHA-256 before offering to launch it. | Only runs when the user installs or updates. |
Machine-readable version: privacy/network-calls.json.
Sandbox surface
Permissions and entitlements
The notarised macOS app uses two hardened-runtime entitlements: audio input and microphone. macOS also requires user-granted Accessibility and Input Monitoring permissions for the hotkey and paste workflow.
The macOS app does not add JIT, unsigned executable memory, library-validation, network-client, contacts, calendar, location, camera, or file-provider entitlements.