REMsound Privacy Policy

Last updated: July 5, 2026

Short Version

REMsound is a local-first Windows dictation app. Core speech-to-text runs on your own computer. REMsound does not upload your microphone audio, does not use analytics, and does not operate a tracking server.

Microphone Audio

REMsound uses your microphone only when you start dictation. Audio is processed locally by the speech model on your computer so the app can type the recognized text into the focused application.

REMsound does not send microphone audio to REMsound servers. There are no REMsound servers for dictation.

Text Reformatting

REMsound can optionally reformat dictated text. Local options use built-in rules or a local tool such as Ollama.

If you choose a cloud provider such as Gemini, Grok, OpenAI, or Claude, REMsound sends only the text you asked to reformat to that provider. Microphone audio is not sent. Your use of a cloud provider is also subject to that provider's own privacy terms.

Local Files

REMsound stores settings and app data on your computer. Depending on your settings, this may include:

config.json for settings, hotkeys, local API keys, and custom shortcuts.
history.jsonl for dictated text history, if history is enabled.
remsound.log for local troubleshooting logs.

You can delete these files from your own machine if you no longer want to keep them.

Downloads And Updates

On first use, REMsound may download the speech model needed for offline transcription. Optional update checks may contact GitHub Releases to see whether a newer public version is available.

These checks are not analytics. They are used to download required app/model files or show that an update exists.

No Analytics Or Advertising

REMsound does not include advertising SDKs, behavioral tracking, or product analytics. It is not designed to collect usage profiles.

Contact

For support or privacy questions, use the support link provided with the REMsound download or Microsoft Store listing.