LingoMitraPrivacy information
Information updated 20 September 2026
Account and learning data
Public courses and prepared opening samples can be viewed without signing in. Guided learning uses Sign in with ChatGPT. The platform supplies a site-specific account identifier, email and, when available, a display name. The app uses the identifier to keep your progress, activity, preferences and requests separate from other learners.
Learning and device data
Server records include guided lesson position, checkpoint feedback, completed parts, streak activity, daily goal, timezone, recent sign-in activity and bounded usage counters. Reader bookmarks, answer drafts, script preferences, playback speed and offline course copies also live in your browser. Browser storage is tied to this device and origin. It does not automatically move between domains or browsers.
Audio and AI
Prepared course narration is shared teaching material. If you choose Record, audio is sent through LingoMitra to OpenAI for transcription when you stop or after 45 seconds. The app does not retain the raw microphone file. Your editable transcript and submitted answer may be sent to OpenAI for reading help or feedback. Feedback is saved with your progress; spoken feedback can be cached privately. Course narration or feedback text may be sent to Sarvam for supported voices or to OpenAI. AI output can be inaccurate.
Optional statistics and conversation services
Usage statistics are off until you accept them. They use pseudonymous browser/session identifiers, coarse device/browser labels, referring hostname, course identifiers and playback measurements. They exclude raw answers and recordings. Cookie settings lets you withdraw without losing lessons. A separately chosen conversation connection sends your conversation or requested answer comparison to its configured provider. Browser speech recognition may send audio to the browser provider.
Service providers
The app runs on ChatGPT Sites with Cloudflare D1 and R2 storage. Requested AI and audio features use OpenAI and Sarvam. A conversation connection uses the provider you select, and browser speech recognition may use your browser vendor. Personal ChatGPT privacy settings do not configure the app’s API accounts. No guarantee of data residency or zero provider retention is made.
Current retention
The app currently has no automatic purge schedule for account learning records, operational records or accepted statistics. Browser records remain until you clear them or the browser evicts them. Provider records and platform backups have separate retention settings which are not controlled by the device-clear button. Responses API requests use store:false; this is not a promise of zero provider retention.
Available controls
Cookie settings lets you reject or withdraw optional statistics without losing lessons. Data controls provides an authenticated account-data download and a separate device export or device clear. Clearing this device does not delete server records or your ChatGPT account. Account deletion and public support/privacy-request handling are not enabled in this release.
About this information
This page explains the implemented data flows and current controls. It is not a completed jurisdiction-specific privacy policy or a claim of legal certification. Avoid including sensitive personal information in learning answers.