I have a large audiobook library made up of ripped Library CDs.
I’m satisfied with the way Gracenote identifies the CDs as I import them into Music.app.
Occasionally it’s not consistent within a book from one CD to another, or the book may be new and not on Gracenote. In those cases I simply select and Get Info and edit in Music metadata info.
Honestly all I need from an audiobook is Title, Author and disk number. I keep it simple.
I also don’t bother keeping but 12 max books in Music.app at any one time. As audiobooks I don’t see the need to have access to all of them.
I took the plunge and now do complete ebook in ABS as well. Kavita reader is better but no metadata (coming 0.7 or 0.8 apparently), but for now I let ABS pull metadata and can’t throw files in folders and don’t have to deal with Calibre! It’s amazing.
Yea, I’m moving from Calibre web. It still is horrible for multi user (read counts don’t stay, it’s really more of a one user interface) and still requires Calibre database. ABS and kavita can use normal files structure.
Edit: Additional Info
Calibre web is a decent looking front end. Unfortunately it is a database structure. It does require a calibre DB to start (But after initial doesn’t require this anymore). It is hard to edit metadata without calibre. Running both calibre and calibre-web can run into database issues with corruption (ask me how I know!). Also there are multi-users, but the read or special columns go across all users, so effectively is only a one user environment. Also if you go to the fire structure and add/remove books without doing it through the calibre-web interface or calibre app itself you will have issues…
Kavita has great readers, syncs progress (doesn’t matter to me since I still download so I can highlight). For PDFs I highlight I can easily just upload back to the native file folder and title with -highlight.pdf to keep both copies. Also Kavita does a great job with multiple users, easier for access, etc. It doesn’t pull metadata (so epub need to have it embedded), and PDF have essentially none. I have issues with this as a front end (but metadata support and scribbling coming). This will be my front end soon.
Audiobookshelf is an audiobook first app, but there is ability to pull metadata (and read books but the readers are horrible), but you can download from it and use as a front end. Also just point it at a multitude of folder structures and it picks it up quickly, and also does a good job with moves/re-titles etc without calibre. Not a database format either. My current solution is to use this for metadata for now in a folder structure than I can use from Kavita or audiobookshelf, and as Kavita gets more improvements will likely move over to this and roll it out to family (right now just using audiobookshelf for family audiobooks.)