Full Stack & AI Developer · HackHarvard
Textify – Digital Humanities Platform
Jan 2024 – May 2024
Digital humanities platform that digitizes manuscripts in Old English, Latin, Greek, and Sanskrit using Google Cloud Vision OCR and advanced computer vision to structure and analyze ancient documents.
Along the way
- Digitized and structured manuscripts across multiple classical languages
- 90–95% text accuracy using Gemini Flash 1.5 and computer vision
- Lexical analysis with CLTK and live English audio with Google TTS
- Streamlined transcription workflow, reducing years of manual scholarly effort
Technology
ReactNext.jsFastAPIPythonGoogle Cloud VisionGoogle TTSCLTKGemini Flash 1.5