translate-book

To translate a book from PDF or EPUB in bulk, you can use a pipeline that converts input into Markdown chunks, translates chunks in parallel using subagents, and merges outputs into multiple formats. This approach handles large-scale translation efficiently while preserving formatting.

Yes, you need Python3, pandoc, and ebook-convert installed to translate EPUB and DOCX files. These dependencies handle input conversion and output generation, while optional Calibre tooling and Python dependencies manage parsing and HTML formatting.

Yes, you can resume a multi-format book translation if the process stops midway by using a manifest-based integrity check system. This validates completed Markdown chunks and allows the parallel translation pipeline to continue from the last processed segment.

The best way to translate a 300-page manual while preserving formatting is to use a chunked parallel processing pipeline. It converts the manual into Markdown chunks, translates them simultaneously with subagents, and merges outputs back into HTML, DOCX, EPUB, and PDF formats.

Yes, parallel subagent translation works with Calibre for output formatting by using Calibre-based tooling during the merge stage. This integration ensures translated Markdown chunks are properly formatted when exported to multi-format outputs like EPUB and DOCX.