Sublern ("we", "our", or "us") operates the Sublern browser extension (the "Extension") available for Google Chrome, Mozilla Firefox, and Microsoft Edge, as well as the Sublern web platform and API accessible at sublern.pro. This Privacy Policy explains what information we collect, how it is used, how your personal data is protected, and your rights regarding account synchronization and learning telemetry.

1. Information We Collect and Process
   1.1 User Account Information
   You can use core video translation features as a guest without creating an account. If you choose to register for a Sublern account (via email and password or Google Sign-In), we collect:

Email Address: Used to uniquely identify your account, enable login, and recover access.
Password (Hashed): If registering via email, passwords are encrypted using one-way bcrypt cryptographic hashing with unique salts. We never store or have access to your plain-text password.
Google Identity (OAuth): When using Google Sign-In, we receive your verified email address and basic profile identifier from Google Identity Services. We do not access your Google contacts, Google Drive, or other account data.
Authentication Tokens: Secure JSON Web Tokens (JWT) issued upon login to authenticate communication between the Extension, web app, and our API (api.sublern.pro).
1.2 Cloud Dictionary & Settings Synchronization
For authenticated users, the Extension automatically synchronizes your vocabulary and preferences with our secure cloud servers:

Personal Dictionary: Words you choose to save, their translated definitions, context sentences captured from subtitles, review counts, repetition intervals, and Spaced Repetition (SRS) learning status.
Extension Preferences: Your customized subtitle styling (font size, colors, opacity, positioning) and feature preferences (auto-pause, speech rate, OCR settings) to ensure a consistent experience across all your devices.
1.3 Learning Statistics & Device Telemetry
To power your personal learning analytics dashboard (/stats) and monitor platform performance, we collect learning activity events:

Anonymous Device Identifier (UUID): Upon first installation, the Extension generates a random, pseudo-anonymous unique identifier stored locally in browser storage. For unregistered users, telemetry is tied strictly to this device identifier without any personal data.
Account Binding: When an existing guest logs into an account, their device identifier is linked to their account so that learning history and statistics are seamlessly preserved in their personal profile.
Subtitle Interactions (word_hover): Recorded when you hover over words in video subtitles to inspect translations, alongside the website domain (e.g. youtube.com, netflix.com, ted.com) where the interaction occurred, allowing you to see your learning distribution across platforms.
Word Saves (word_save): Recorded when you add words to your dictionary, including the source website domain.
Trainer Sessions (trainer_start): Recorded when you initiate flashcard practice sessions and the selected trainer mode (Spaced Repetition SRS or Scramble anagrams).
Subtitle Modes (subtitle_use): Recorded when subtitles are activated and which mode is used (Native player subtitles, Video OCR, or Audio OCR).
What we NEVER track: We do not log your browsing history, unrelated website visits, search engine queries, keystrokes outside subtitles, private communications, or video watch history. We only track interactions with subtitle elements on pages where you explicitly activate Sublern.
1.4 Transient Media Processing (Zero Permanent Storage)
Certain advanced AI features require temporary transmission of media directly to specialized processing providers. This data is processed transiently in real time and is never stored permanently on our servers:

Audio Capture (Audio OCR / Whisper): When you activate audio subtitle generation, audio from the active video tab is streamed via encrypted WebSocket to Deepgram API for speech-to-text conversion. The audio stream is processed in-flight and discarded immediately.
Video Screen Capture (Video OCR): When you enable recognition of burned-in video subtitles, a cropped image containing only the subtitle portion of the video player is captured and sent to OCR.space API for text recognition. Images are processed transiently and immediately discarded.
Subtitle Word Translation: Subtitle words and their sentence context are sent to Reverso API to retrieve translations, parts of speech, and bilingual examples. No user identifier is attached to translation queries.
2. How We Use Information
We use the information we collect strictly for the following purposes:

Delivering Core Features: Translating subtitles, pronouncing words, and generating real-time transcriptions.
Cross-Device Vocabulary Sync: Synchronizing your personal dictionary and customized extension settings across all browsers and devices where you sign in.
Personal Learning Dashboard: Providing you with insights into your study habits, daily activity timelines, vocabulary expansion rate, and flashcard trainer accuracy.
Security & Fair Use: Authenticating users, securing API endpoints, preventing unauthorized access, and enforcing fair use limits on high-cost AI services (Deepgram speech recognition and OCR).
Aggregated Platform Reliability: Monitoring system uptime, error rates, and feature usage to maintain high availability and troubleshoot issues.
No Advertising or Data Selling: We do not sell, rent, or trade your personal information to third parties. We do not use your data for advertising, behavioral profiling, or marketing purposes.

1. Third-Party Service Providers
   We integrate with trusted third-party service providers solely to perform specific technical operations on our behalf:

Reverso API (api.reverso.net)
Provides contextual word translations and usage examples for video subtitles.

Deepgram API (api.deepgram.com)
Provides real-time speech-to-text audio transcription for videos lacking native subtitles.

OCR.space API (api.ocr.space)
Performs optical character recognition on hardcoded / burned-in video subtitles.

Google Identity Services
Provides optional, secure single-sign-on (SSO) authentication for user accounts.

1. Data Security and Storage
   We implement rigorous administrative, technical, and physical security measures to protect your information:

Encryption in Transit: All data exchanged between your browser, extension, and our API servers is encrypted using modern TLS (HTTPS) and Secure WebSockets (WSS).
Secure Credential Storage: Passwords are never stored in plain text and are hashed using salted bcrypt cryptography. JWT tokens are cryptographically signed and stored in protected browser extension storage.
Infrastructure Protection: Our backend databases and servers are hosted in secure enterprise datacenters with firewall isolation, continuous monitoring, and strict principle-of-least-privilege access controls.
5. Your Rights & Data Control
You retain complete ownership and control over your vocabulary and account data:

Data Export: You can export your entire personal vocabulary at any time from the dictionary settings in CSV, JSON, TXT, or Excel formats.
Vocabulary Deletion: You can delete individual words or clear your entire dictionary at any time directly through the extension or web interface.
Account & Data Deletion: You have the right to request full and permanent deletion of your account, vocabulary, and all associated statistics. To do so, contact us at korol20041@gmail.com.
Uninstalling the Extension: Removing the Sublern extension from your browser instantly wipes all locally cached settings and offline dictionary records.
6. Browser Extension Permissions
To function properly across video platforms, the Extension requests the following browser permissions:

Storage (storage): Used to persist your local preferences, offline dictionary words, and active session tokens.
Active Tab & Tabs (tabs): Used to interact with video subtitles on the page you are viewing and communicate with our web authentication bridge when signing in.
Host Permissions: Required to communicate with Sublern servers (api.sublern.pro, sublern.pro) and third-party APIs (Reverso, OCR.space, Deepgram).
7. Changes to This Privacy Policy
We may update this Privacy Policy from time to time to reflect changes in our services, security practices, or legal requirements. When updates are published, the "Last updated" date at the top of this page will be revised. Continued use of Sublern after revisions indicates your acceptance of the updated policy.

1. Contact Information
   If you have any questions, concerns, or requests regarding this Privacy Policy or your personal data, please contact the developer directly at: korol20041@gmail.com