Batch Book Translation & OCR

This skill automates the complex workflow of digitizing and translating historical volumes. It manages a multi-stage pipeline that first splits two-page spreads into individual images, performs high-accuracy OCR using Gemini vision models, and finally translates the text while maintaining context across page breaks. Designed for scholarly and archival projects, it leverages the Gemini Batch API to reduce costs by 50% while producing semantically tagged Markdown output that preserves the structure and metadata of the original work.

主な機能

ユースケース