Roland0 wrote:
>
> How well this will work will depend on how accurate the models are, how
> exhaustively a track is characterized by the combination of all classes,
> and on how well the models are a fit for the tracks. For moods etc. the
> latter will be much less of an issue then for genres. For instance, none
> of the genre models work that well for my collection, most likely
> because they are trained on representative music collections and thus
> contain classes not applicable to my tracks (i.e. genres that are not in
> my collection), and also because they are too coarse-grained for my
> taste ("rock" or "pop").
Yeah, I've mainly been using the 'moods' (and bpm) as I'm not too
interested in the genres. I had a play with selecting a track, and then
getting the %difference in all moods with other tracks, and using this
as a similarity. This -seems- to work quite well for a selection of
~1600 tracks. With that in mind, I'm going to analyse ~20k tracks (which
will take some time, I'm averaging 14 tracks/min) and see if this
approach works.
I'm wondering, seeing as I only really care about similarity, whether I
could just use the low-level results - and compare (a selection of)
these against files to get a similarity. I'm saving the raw JSON output
from the high-level analysis, so trying with low level analysis will not
need a complete re-scan. I just need to decide which of the values to
use...
*Material debug:* 1. Launch via http: //SERVER:9000/material/?debug=json
(Use http: //SERVER:9000/material/?debug=json,cometd to also see update
messages, e.g. play queue) 2. Open browser's developer tools 3. Open
console tab in developer tools 4. REQ/RESP messages sent to/from LMS
will be logged here.
------------------------------------------------------------------------
cpd73's Profile: http://forums.slimdevices.com/member.php?userid=66686
View this thread: http://forums.slimdevices.com/showthread.php?t=108278
_______________________________________________
plugins mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/plugins
