Capability matrix
The operating modes against what they run on, and every input format against the helper it needs - if it needs one. The page is generated from the app’s own facts and is held against them by the build’s checks.
# Operating modes and their dependencies
| Mode | What it is for | What it does | What it runs on |
|---|---|---|---|
| Desktop GUI | Everyday conversion on Windows | Seven-page side menu, F1 help for the part of the window you are in, text and interface size, themes, drag-and-drop, a conversion queue. | Windows 10/11 (64-bit). Nothing to install beyond the app itself: pdftotext ships with it; the optional helpers below are used only when a file needs them. |
| Command line | Repeated tasks, scripts, unattended runs | All conversion options as flags, custom output folders, engine selection, exit codes for scripts. | The doc-html-translate executable; no runtime install for the standard text and e-book formats; the same optional helpers on demand. |
| Browser extension | Reading in Chrome or Edge | Converts a document right in a tab and translates the page with the browser's own translation; recognises text on page images (OCR). | A Chromium browser (Chrome or Edge); the read access to local files for documents opened from disk. Nothing is installed. |
| Translation engines (optional) | Translating the converted text | The browser's own full-page translation (free, no key); a local Ollama model; Google Cloud Translation with a spending guard. | Page translation needs nothing extra; Ollama needs a local server on localhost:11434; Google Cloud needs your own API key, and -max-cost stops a request before it is sent when its estimate exceeds the limit. |
# Formats and the helpers they need
| Format | File extensions | What it needs |
|---|---|---|
| EPUB e-books | .epub | Nothing - native parser. |
| PDF documents | .pdf | Nothing on Windows - pdftotext ships with the app; elsewhere an installed Poppler; a built-in reader as the fallback. |
| Kindle books | .mobi, .azw3 | Calibre, installed (non-DRM files only). |
| FictionBook | .fb2 | Nothing - native parser. |
| Rich and plain text | .rtf, .txt, .md, .html | Nothing - native parsers. |
| Comics (zip and tar archives) | .cbz, .cbt | Nothing - unpacked by the app. |
| Comics (rar and 7z archives) | .cbr, .cb7 | 7-Zip, installed. |
| Pictures and scans | .png, .jpg, .jpeg, .webp, .gif, .bmp, .tif, .tiff | Optional: the external Tesseract engine for OCR; English data ships with the app, other languages download on demand. Without it the pictures are kept without text plates. |
| JPEG XL pictures | .jxl | Nothing - decoded in-process by the decoder bundled with the app. |
| FileDO secret files | .fd-sec | The FileDO utility, installed, and the container's password; the decrypted copy is deleted as soon as it has been read. |
Good to know
OCR and the optional helpers are best-effort: a missing Tesseract or helper never stops a conversion - the step that needed it falls back and says so. DRM-protected files are not supported.