Urgent: Metadata Corruption Patterns Found Across My Library (Filename/Audio Intact)

Please fill in this bug report template:

Lexicon version: …
Latest Version of Lexicon: Version 1.9.6

Operating system (remove one): Windows / Mac
MacOS M4 and Windows 11

Hi Christiaan. I hope you are well. I am running into a major issue in my library and wanted your insight on the best way to detect and fix it, whether through Lexicon or another tool you recommend.

My library is managed through Lexicon and sent to Engine DJ for performance. While reviewing some of my live gig playlists, I found many tracks where the metadata in Lexicon no longer matches the actual audio file, even though the filename and the audio played are always correct. Something altered the metadata at some point, because during the original live performances the displayed track titles matched the files and the correct songs played. The incorrect metadata also shows up in Engine DJ, so whatever is happening is being carried over and is not limited to Lexicon.

In the example shown in my screenshots, the playlist row displays the title “This Is How We Do It (Clean),” but when opening the track, the filename correctly shows “California Love” and the audio played is California Love. In this particular case the artist, album, and artwork appear correct, but the title metadata is wrong. In other examples I found, the title, artist, album, and additional metadata fields are all incorrect. The only consistent piece is that the filename is correct and the audio played always matches the filename.

I first noticed this about one month ago during a gig on my Prime 4+. I cued up a track expecting one song, but the displayed title did not match the audio that was about to play. That is when I realized something was wrong. Since then I have seen similar issues across both older and newer playlists, so it is not isolated to one playlist or timeframe.

Right now I am checking each track manually by comparing the Title field to the filename in the Location field, which is extremely slow with a large library, and I am repairing each mistake as I find it.

I searched the forum but did not find anything describing this behavior. I am hoping there is a faster way to identify these mismatches. For example, is there any way in Lexicon to create a smart playlist or run a process that compares the Title metadata against the filename and returns all tracks where none of the words from the Title appear in the filename? Even a partial text-match comparison would make cleanup much easier. If Lexicon cannot do this, is there another tool you recommend (such as mp3tag or another batch comparison utility) that can scan the library, compare metadata to filenames, and output a list of inconsistencies to fix?

I will keep reviewing and repairing things manually until I hear back. Any guidance or ideas you can share would be greatly appreciated. This is my life blood and I want to fix this as quickly as possible.

Thanks Christiaan,

Joe

Screenshot:

Are you writing tags in Lexicon? Because if you only sync to Engine then it won’t update ID3 tags. Engine reads the info from its database and so does Lexicon.
If you never write tags, it will have old info in there. Engine might be using that (when you reload tags in there)

Hmm I normally write/update tags in Lexicon but it doesn’t mean I’ve never updated in Engine in the past; been using both for a loooooong time. Just noticed this situation though. Right now MP3Tag and its filtering is helping me identify the false metadata by checking to see if the “track title name” exists in the filename. It’s actually saving me time but as far as how it happened man I wish I knew.

I’ve never heard of Lexicon writing the wrong titles to files though. Not saying it’s impossible but I think it would have been reported earlier since that would be pretty serious. Unless it’s a super rare occurrence somehow (not unheard of in programming)