Privacy policy

Last updated August 20, 2026

Who we are

Red Flow is a dictation app for Yiddish. It records your voice when you ask it to, converts it to text, and types that text into the app you are using. This policy explains exactly what we collect, why, and how to remove it.

What we collect

  • Account information. Your email address, your display name, and a securely hashed password. We never store your password in readable form.
  • Voice recordings. Audio is captured only while you hold the dictation shortcut, hold the microphone button, or have locked a hands-free session. The recording is sent to our servers to be transcribed and is stored so that transcription quality can be improved.
  • Transcripts. The text produced from your speech, along with the length of the recording and which engine produced it.
  • Corrections. If you edit a transcript, we store the corrected version alongside the original so the speech model can learn from the difference.
  • Your dictionary and snippets. Words, names and phrases you add.
  • Devices. A device name and platform for each signed-in session, so you can see and revoke your own sessions.

What we do not collect

  • We do not read, collect, or store the contents of your screen.
  • The Android accessibility service is used only to show the microphone bubble when a text field is focused and to insert your dictated text into that field. It never activates on password fields.
  • We do not sell your data, and we do not use it for advertising.
  • We do not record audio in the background. Recording only happens when you start it.

Why we keep recordings

Yiddish speech recognition is a small field with limited training data. Keeping your recordings and any corrections you make allows the model to improve at recognising Yiddish, including names and words specific to how you speak. Recordings are associated with your account and are not published or shared with third parties.

Who processes your data

  • Railway hosts our servers and database.
  • RunPod runs the GPU speech recognition service that transcribes audio for accounts on the accelerated tier.
  • ivrit.ai produces the open speech recognition model we run. Your audio is processed by that model on infrastructure we control.

These providers process data on our behalf and are not permitted to use it for their own purposes.

Deleting your data

  • Delete a single dictation. Open the history list in the desktop app and delete the entry. Its audio is deleted with it.
  • Delete all dictations. Web portal, Settings, Clear all dictations.
  • Delete your entire account. In the Android app open Home and tap Delete account, or visit redflowapp.com/delete-account. This permanently removes your account, transcripts, recordings, dictionary and snippets. It cannot be undone.

Security

All traffic between the apps and our servers is encrypted with HTTPS. Passwords are hashed with Argon2. Access tokens are short lived and refresh tokens rotate on every use.

Children

Red Flow is not directed at children under 13 and we do not knowingly collect their data.

Changes

If this policy changes in a way that affects how your data is used, we will update the date at the top of this page.

Contact

Questions, requests, or data concerns: support@redflowapp.com.