I think I'll put a page on the wiki, probably called "Classical Music
tagging: the maniac method".

I don't think the whole method/scripts can be used "as is" by someone
(the "everyone has is own graal" problem) but for sure this could be a
source of inspiration for others.

I'm in the way to finalize everything from tag fields to mp3tags script
so I won't put it in details yet, but here is the idea:

1. Get raw information from online music databases (i have a preference
for amg for its consistency). I prefix those tags with the source name
like amg-* or arkiv-*

2. Extract what you want in personal tags with scripts. I prefix those
tags with my-*

3. Enrich my-* tags with missing information

4. In the very end, extract from the whole stuff the standard ARTIST,
ALBUM, ALBUMARTIST, COMPOSER, CONDUCTOR, YEAR, ORIGYEAR with special
scripts that choose the best tags from the previous tags.

This way, if I want to switch my whole collection from Slimserver to
itunes or windows media player, I just have to rewrite scripts used in
step 4 and regenerate standard tags from the detailed ones.

Here is an example with this disc:
http://www.arkivmusic.com/classical/album.jsp?album_id=100617


1. First, I look in the web for the AMG reference and I use a first
mp3tag source to extract:

amg-album-duration=65:56
amg-album-featuredartist-name=Giuliano Carmignola
amg-album-featuredartist-sql=41:95916
amg-album-genre=Concerto
amg-album-id=W   124562
amg-album-label-n-ref=Archiv[002894748952]
amg-album-perform-rating=4 Stars
amg-album-releasedate=2005
amg-album-sql=43:124562
amg-album-title=Concerto Veneziano [Hybrid SACD]
amg-work-genre=Concerto
amg-track-work-sql=42:6465
amg-track-number=1
amg-track-composer-name=Antonio Vivaldi
amg-track-title=Double Orchestra Concerto, for violin, 2 string
orchestras & continuo in B flat major, RV 583
amg-track-title-detail=1. Largo e spiccato - Allegro non molto
amg-track-length=05:20

Than another script to get the credits:
amg-album-credits-withrole=Andrea Marcon/Conductor;Giuliano
Carmignola/Violin;Venice Baroque Orchestra/Orchestra;Marita
Prohmann/Executive Producer;Jurgen Bulgrin/Engineer;Daniel
Fesquet/Booklet Editor;Ingmar Haas/Editing;Merle Kersten/Art
Direction;Reinhard Luthje/Liner Note Translation;Arend
Prohmann/Producer;Stewart Spencer/Liner Note Translation;Roger-Claude
Travers/Liner Notes;Ulrich Vette/Balance Engineer, Surround Mix;

Than another script to get the detail of the "work" piece:
amg-work-title=Double Orchestra Concerto, for violin, 2 string
orchestras & continuo in B flat major, RV 583
amg-work-composer-sql=41:8090
amg-work-composer-name=Antonio Vivaldi
amg-work-average-duration=13:10
amg-work-worktype=Concerto for Orchestra
amg-work-composition-date=before 1742
amg-work-id=C     6465



Every script keeps track of the amg "sql" and "id" for hypothetical
future processing/links

2. From this, I extract my personnal fields from the previous one with
another script to get (including catalog/opus):

MY-COMPOSITIONYEAR=1742
MY-LABEL=Archiv
MY-ORCHESTRAS=Venice Baroque Orchestra
MY-INSTRUMENTISTS=Carmignola, Giuliano
MY-CATALOG-NUMBER=RV 583

3. I manually enrich what is needed to be overriden (if i don't like
amg-album-title for example)

my-rec-date=2004
MY-ALBUM-TITLE=Concerto Veneziano
my-genre=Classical;Baroque;Violon;Ensemble à
cordes;Cl18;Concertos;Clavecin;Luth;Viole

4. I use another script to generate the standards tags with reformated
(for example for alpha sorting):

Album=Concerto Veneziano - Carmignola - Archiv
Tracknumber=1
TITLE=Double Orchestra Concerto, for violin, 2 string orchestras &
continuo in B flat major, RV 583 - 1. Largo e spiccato - Allegro non
molto
COMPOSER=Vivaldi, Antonio
ARTIST=Vivaldi, Antonio
CONDUCTOR=Marcon, Andrea
ALBUMARTIST=Venice Baroque Orchestra
YEAR=2004


The only manual and optional part is step 3, but in reality it's a
little more difficult as sometimes albums are difficult to find,
sometime the disc doesn't exist so I have to enter some data manually.
But in the end, the result worth it. And you don't have to be as maniac
as I am ;).

If I change my mind or my program, all I have to do is rewrite script
of step 4 (the more simple) and run it again on my whole library.

I keep you informed with this if you're still interested or if others
are.


For your information, I attached the script I use to extract
opus/catalog names.


+-------------------------------------------------------------------+
|Filename: Extract  Catalogue.txt                                   |
|Download: http://forums.slimdevices.com/attachment.php?attachmentid=3105|
+-------------------------------------------------------------------+

-- 
vrobin
------------------------------------------------------------------------
vrobin's Profile: http://forums.slimdevices.com/member.php?userid=11705
View this thread: http://forums.slimdevices.com/showthread.php?t=37268

_______________________________________________
ripping mailing list
[email protected]
http://lists.slimdevices.com/lists/listinfo/ripping

Reply via email to