The track titles in that screenshot are the track titles on Musicbrainz (I checked). They’re not the tags I use because mine have hyphens, not colons. I assume that whether Plex uses Musicbrainz’s track titles or your own depends on if you set the library to prefer local tags (which I did.)
Musicbrainz says they’re particular about the accuracy of a release’s entry mirroring the release itself (I know because I asked them about custom artwork), so I guess if the audiobook was published with marks separated by timecode rather than by chapter, then that’s the way Musicbrainz prefers it.
Thanks. I actually have several books in my Audible account that Audible no longer has a listing for. I found out when I was trying to add The Wonderful Wizard of Oz read by Brooke Shields (who does a better job than Anne Hathaway, imo) to Musicbrainz that I could no longer use the Audible store page to help me populate it. For this reason, I find using Audible as a source for metadata to be sometimes lacking.
I’ve looked through your entries for the ExForce series. It’s all very thorough. You’ve done a great job with it. It’s interesting to see how something like Homefront is tagged differently as they have tags for Audiobooks vs Audio dramas.
This is actually the biggest reason I can think of for trying to do something more dedicated for audiobook metadata. Something equivalent to thetvdb and tmdb.com but just for audiobooks. I’ve tried to think of different ways to populate it without relying solely on user contributed content. I wondered if i could make it act as a “go-between” that cached data from audible but also was editable. So if a book disappeared, the data remained accessible. And you could add new entries as well. Unfortunately, my work projects have picked up to the point that I only really have time to keep stuff working rather than expanding what I’ve already got. And I can’t imagine what kind of nightmare curating the database would be.
I’ve tried looking into pulling from goodreads, but their standard API is a bit lacking when it comes to trying to pull format specific information. And I don’t think that making every user apply for their own developer api key is the right answer. But I haven’t given up hope just yet.
Truth be told, that’s one of two primary reasons I’m growing my audiobook library from Audible rather than from torrents. (The other reason being that I like the idea of the author and/or publisher getting paid.)
I’ve acquired more than enough audiobooks through file-sharing that are cut along time stamps, or split into equal-filesized chunks—and especially enough that sound like someone ripped them either from audio cassette or scratched discs borrowed from the public library—that I generally don’t trust that whole enterprise anymore. Plus, my reencoding dumpy files in an effort to chapterize them properly just reduces their audio quality that much more. And this is coming from someone who reencodes audiobooks using mp3 VBR 8 Mono (or VBR 7 Stereo if it’s an audio drama).
@macr0dev - Just wanted to say nice work on the plugin. Is there any ETA on when the update is coming?
@chyron8472 - Nice work on giving another option. I was also looking into MusicBrainz for this - but wasn’t sure how complete that DB was for Audiobooks. I may give that a try as well.
thank you @macr0dev and @chyron8472.
I’m always interested in “background stories and how things work” and its very interesting to read your posts here!
You’re a big support for the community!
Hi macr0dev, there is a new beta version of plex (released 3 days ago). It still doesn‘t seem to have a fix for this. If plex is unable or unwilling to fix the problem … would it be possible to implement a workaround? There is a fork on github that checks if the title attribute exists (and assigns the value of the name attribute to it if it doesn‘t if I understood it correctly). Would that help to mitigate the problem?
@natom23, when the problem first appeared I spent a day or two looking at ways to work around the problem. At best I could make it work when a book was added, but if it matched incorrectly there was no way to correct it. Not passing along the manual search variable is really killing things. I had hopes that this would be just a temporary problem, and when it was suggested that it may just be a few weeks, I didn’t want to get temporary work around code out into the wild just for it to get fixed… I may have to revisit that idea.
Should this metadata agent be working now? There’s been at least one beta update since the last official message (the one I installed 5 minutes ago) and the audiobook metadata agent is still not working.
Admittedly it could be something I’m doing wrong. Any confirmation either way?
FYI, it looks like this may have been fixed (beta channel PMS version 1.18.4.2164)! I was able to use the Audiobooks agent to retrieve metadata for a book that I haven’t been able to add since this broke.
