Total New-Beee

[HereNow]

Hi.

I’ve had this collection of music that has been driving me crazy for ever. Mostly tech-house trance, hard to find and awesome. I actually have a lot of the vinyl Discogs lists. I have been experimenting with Jaikoz.

I find that, even though the titles (filenames) are very complete, (spent days doing that years ago) the actual positive hits on the tracks I have is less than 50%. (for fill of meta data (which they all mostly are missing))

Is there anything I can do to increase that success rate? I was hoping to get a high percentage of success so it won’t take me forever to fill in the missing blanks myself.

I understand that there are limitations, and the fact the software does what it does at all is… just great! But since these are mostly singles it seems the success rate is low. (even if the file (as in title/artist) names are correct in the file name of the track.

Any suggestions? I am new to this and just started experimenting with the trial, but if I can tweak it somewhat to be more successful I would totally purchase it as the time saving alone is worth it. If not, I’ll just keep on going by file name. (But as you can imagine, that sucks)

Anyways, Thanks.

PS. What is the purpose of SongKong?

[paultaylor]

Okay, first of all Jaikoz was developed a few years back the idea being it gave you all the tools you need to organize your music using a combination of automated matching and manual editing. But although it is powerful it can still be time consuming once you start tinkering.

SongKong was developed just last year the aim is to provide automated matching with just a couple of clicks, it builds on the ideas of Jaikoz and learns from a few mistakes made in the design of Jaikoz. For example Jaikoz needs more memory to fix more songs, SongKong does not.

Generally I would recommend potential customers look at SongKong before Jaikoz because it currently is better at automated matching and simpler and easier to use. Then Jaikoz works well as swiss-army knife solution for difficult problems to crack.

Now both applications are release based, and in the first instance group releases by their folder (i.e one folder = one album) so I ask is this how your songs are organized.

Secondly both use MusicBrainz and Discogs database, but recent changes made by Discogs have reduced the effectiveness of Discogs matching. I am currently working on hosting a copy of the Discogs database myself and tweaking the search so that it works much better for music tagging , this is going very well but it will probably be a month before it is released. Because Discogs is the better database for electronic music this accounts for the less than steller results you are seeing at the moment, but if you rerun the matching a few times this should increase the num,ber of songs matched.

[paultaylor]

Local version of Discogs server is now in use.