If your export is in a CSV-like format, I've been working with some great tools that simplify editing and automated transforming/correcting/cleaning up the data when it's in this format.
So far I can highly recommend as the two most valuable "CSVed" and CSVfix".
===============
total side tangent, just in case it's of interest to anyone feel free to contact me for further details
I've been constructing a database model, based mostly on the MusicBrainz schema and using Data Crow, that allows me to import my tag data into the proper separate relational tables, e.g.
- Genre
- People (AlbumArtists, other performers, composers/lyricists, engineer/producers etc.)
- the main Genre tag is actually an attribute of the artist
- "Event" - either a concert date or a MB releaseGroup
- Release
- Work (abstract object for songs)
- Recordings (tracks)
- using ContentGroup for track-specific genre classification
- Source (only level related to actual folders/files)
I'm now working on going the other direction, using relational "joins" on the CSV exports from these various tables in order to re-create a single table suitable for MP3Tag importing back into audio file tags.
Later I'll look at outputing data suitable for MediaMonkey, iTunes etc
This will allow me complete control over my own data without depending on the fragile data store of the audio file tags themselves as the primary store, nor depending on proprietary cataloging apps that don't respect their users' control over their metadata and tag integrity.