BlogJuly 2, 2026
A dictation app's microphone is active every time you use it, which means it hears literally everything you say while it's on: drafts, private notes, things you'd never type into a search bar. Being careful about what happens to that audio isn't paranoia. It's a reasonable question to ask before you install anything that listens.
Converting speech to text requires sending the audio somewhere for transcription, whether that's an on-device model or a cloud service. There's no version of dictation software that skips this step entirely. So the useful question isn't "does it process my audio," since it has to. It's what happens to that audio once transcription finishes: is it kept, or is it gone.
Your audio is sent over an encrypted HTTPS connection to be transcribed and cleaned up. It exists in memory only for the length of that request. It is never written to disk, never saved to a database, and never used after the text comes back and lands at your cursor.
To be equally clear about the other side of this: the text Smpliflow produces, not the audio, is kept for your own convenience, capped at your last 50 dictations, so the app can show you a history of what you've said. That's a genuine trade-off, not a loophole, and it's worth naming plainly rather than letting "we don't store your audio" imply that nothing at all is retained.
The full breakdown, including what's covered and how to request deletion, is in the privacy policy.
You can dictate a draft email, a rough work note, or something you're still figuring out how to phrase without a recording of your voice existing anywhere afterward. What's retained is only the text you already chose to put on your screen, not the audio that produced it.