Executive Summary

Users cannot extract burned-in subtitles from YouTube, Instagram and TikTok to practice speaking. A Chrome extension captures on-screen subtitles in real time, OCRs them, translates and outputs pinyin and exportable flashcards. Improvements in lightweight browser OCR and client-side ML allow on-device or low-latency transcription without sending full video to servers, addressing privacy and platform policy concerns. Short-form video consumption (TikTok, Reels) has grown rapidly, increasing frequency of the pain - the source user watches these as a daily practice. Paid workaround signal in the source suggests users will pay for a reliable tool. Also, modern extension APIs and increasing demand for authentic-content language immersion make this the right time to ship a browser extension that integrates OCR, transliteration and export workflows. Focus on real-time, frame-accurate capture plus language-aware postprocessing: OCR tuned for subtitle fonts and video compression artifacts, Chinese-specific normalization and pinyin transliteration, and export to SRS/Anki. The source user explicitly said "subtitles were ingrained into the video and not provided as captions" and had tried other extensions without success, indicating a gap for an extension that pairs robust subtitle OCR with transliteration and study exports. Collecting user corrections and aligned timestamped pairs can create a small proprietary corpus of burned-in subtitle OCR corrections to improve accuracy over time.

Market Opportunity

Capture baked-in video subtitles and transliterate to pinyin via chrome extension targets a $3.0B = 50M users (people actively using online video to learn languages globally) x $60 ARPU annually total addressable market with low saturation and a year-over-year growth rate of 8-12% CAGR for digital language-learning tools and add-on learning utilities.

Key trends driving demand: Short-form video growth -- more language content and embedded subtitles appear on TikTok/Reels increasing exposure opportunities for learners; Consumer ML at edge -- browser and on-device models reduce latency and privacy concerns for OCR/translation; Self-directed learning rise -- more learners use authentic content to improve fluency, creating demand for study tooling; Cross-platform content consumption -- learners move between YouTube, Instagram, TikTok, increasing need for a unified capture tool.

Key competitors include Language Reactor (formerly Language Learning with Netflix), Copyfish OCR, Substital / Subtitle add-on extensions, Workarounds: Google Lens / manual screenshot + Google Translate.

Sign in to access

Capture baked-in video subtitles and transliterate to pinyin via chrome extension

Executive Summary

Market Validation

Market Opportunity

More in EdTech

Writers waste time formatting/verifying citations — AI automates sourcing & formatting

Reference librarians forced to be 'AI experts' — AI‑augmented reference platform

Students constantly relearn tools; unify electronics learning with a lifelong PCB workflow

Frustrating SQL tutorials — teach applied SQL with realistic bank scenarios

Bulk digital certificate issuance — automated, verifiable credential pipelines

Mock-trial pain: practice with AI role players, no full cast needed