Another week, another set of features added to Calibre, the open-source ebook reader, converter and manager.

Calibre 9.18 adds more accessibility tools to its editor, with Tools > Accessibility now letting you set the language for books and HTML files. Image descriptions, including those of inline SVGs, can be edited and EPUB Accessibility metadata modified.

The book check feature now catches common accessibility problems, including images that lack descriptions and skipped heading levels, and flags elements with an epub:type but no matching ARIA role. Helpfully, it can automatically fix many of the issues it finds.

New ebook sources

And Calibre’s now includes three additional sources for finding content.

Standard Ebooks offers public-domain books with, helpfully, improved formatting for ebooks readers (some older titles found on places like the Gutenberg project can be poorly laid out).

You can also ferret through the annuals of the Directory of Open Access Books (DOAB), which offers scholarly books that are free to read, and the Directory of Open Access Journals (DOAJ), which surfaces open-access academic journals.

Content Server updates

If you use KOReader (a more capable, open-source ebook viewer apps that can be added to many ebook devices, including Amazon Kindle), you can now sync reading position via the Calibre content server (accessed in a web browser).

First set the custom sync server in KOReader to http://server:port/kosync, and then sign in with your content server user account details.

This update adds other improvements to the content server. It can now serve OPDS 2.0 feeds, stream CBZ comics (with correct manga reading direction) and lets you search book contents using Calibre’s full-text index.

Other changes in Calibre 9.18:

  • E-book viewer now lets you limit search to the current chapter
  • New search date filters, including lastmonth, thisyear & monthsago
  • Interrupted voice (read aloud) model download can be resumed
  • New calibredb commands to manage notes, annotations and more

Fixes include:

  • Better page number detection and handling of large images in PDFs
  • TXT and RTF output and input fixes
  • Errors when reading some damaged EPUB files were resolved
  • Converting very small PDF files no longer causes a crash
  • Sorting merged highlights by position in the book
  • AI game fix for generating ‘worlds’ with Gemini

Plus, there are the usual tweaks to better scrape news sources.

The v9.18 update fixes quirks fetching content from the websites of British newspaper The Times, Sunday Times and Financial Times, as well as The Boston Globe. It also betters its handling of content from the den of nerd debate, Hacker News.

Install Calibre 9.18

You can download Calibre from its official website or get the source code from GitHub.

On Linux, Calibre is installed and updated via a command-line script. It downloads and moves the binary to a system location, then generates a .desktop shortcut to enable you to find and open it from the app launcher, context menus, etc.

You can also get Calibre on Flathub or use the Calibre snap. But, as neither of those builds are verified or official, any bugs or issues you encounter while using them should be reported to the package maintainer rather than upstream.