#musicbrainz

if I didn't look closely, I would've thought that my ghost was over here talking ;) Gilou

luks: but yeah, with those tracks I can't even tell them apart most of the time, particularly in the unison parts :/

doesn't help that one of the 3 people on tracks 1 & 3 is the same person playing two characters

I think https://acoustid.org/fingerprint/20186885/compa... is comparing one of the vocal versions to the karaoke track. all the differences are in the sections with vocal parts.

gcilou, hehe :) sorry...

;)

Freso, well, if it feeds the files with the mbid, I'm golden, I can just use azbsubmit anyway

abzsubmit* damn that's hard

:p

also it's quite frustrating that beets' binary is called beet.. had me wondering if the install failed for a while huhu

Gilou: You'll have to complain to sampsyo about that. :p

beets has its own IRC channel at #beets btw, if you need beets specific help. :)

I will, I have plenty of talking to do with him (assuming he's a he, and he is the beets dev :))

yep

Is "expand/collapse release groups" user script broken for others?

Gilou: Also, you could just make an alias for it.

hehe indeed, it's what I did immediately.. but it was indeed confusing ;))

gcilou: I actually thought at first you decided to be a bit less "gci" since it's no longer gci time :p

(but then I realised it was probably someone else)

Yeah I was going to change it but that sort of seems counterproductive now.. Just pretend that gci stands for something else ;)

great coders inc. lou? :p

Sure:)

gcilou: If you're going to keep that nick, you should register it with nickserv :)

/query nickserv help register

hoa!

+W

https://freenode.net/ has changed!

Yeah okay :). I'll do it later. Taking a math test now :)

kloeri: You should warn people about such things! I near had a heart attack!

(Not really, but...)

gcilou: https://freenode.net/kb/answer/registration

Freso: we did! there was a blog post and global notices and stuff! :)

Psh.

You didn't message him personally by pigeon!

^

well, that is quite rude of me

I shall endeavour to do so the next time

👍

Ooh. https://freenode.net/news/new-extban-j

kloeri: Unrelatedly, is there a reason eir isn't +v'ing me in #freenode?

it's manually triggered unless you've been voiced within the last 24 hours

and there you go

alright! Nick is registered

🙌

gcilou: Want a cloak to go with that new nick?

(And if so, what's your username on mb?)

cloak? mb username = caroline-g

gcilou: Try and do /whois Freso

That part after the @ is my clock

*cloak

oh ok

what does mine say now?

nvmd I see

Freso, and sure I'll take one ;)

Freso likes giving out cloaks.

He must think it's always cold

Or he just wants them to be fashionable.

gcilou: There's a reason I'm moving to Barcelona.

:)

the paella?

Yes.

Vamos a la paella!

They need cloaks in BCN? Anti-pickpocket cloaks, probably.

Freso: do I need to do anything?

gcilou: Nope.

okee

I am cloaked :)

🙌

yay for cloaking devices

Stealth boys galore!

http://memory-alpha.wikia.com/wiki/Cloaking_device, of course

kepstin: I must say, finding differences in those songs is properly hard, even after having the exact times from the matching algorithm :)

heh, and you don't want to make the matching oversensitive, since that'll make it throw away matches for things that are actually the same :)

and the new matching algorithm will depend on <5s segments that are distinct enough to tell those songs apart, so it won't be very robust :)

I'm not working with averages over the entire fingerprint this time, so it shouldn't be a problem, I basically split the fingerprints into segments and if there are some segments that do not match well, it will not consider those two fingerprints to form the same "acoustid"

luks: Will there be some kind of versioning so it's possible to tell when generation an AcoustID is?

probably not, this won't make it to acoustid for a long time and when it does, it will only affect a very small chunk of the database

Clint: paella is from Valencia! :p

so, it's sort of like it'll do the average matching to find match candidates, then if it's a case where some small segments are very different and the rest is the same, it'll throw away the match?

well, it will still give you the match when searching, but it will be more strict about merging individual fingerprints into "acoustids"

ah, so since it'll keep them separate, you'll get a bunch of acoustids with different match% returned, instead of only one acoustid back?

you already get multiple acoustids back with some score

so everything more or less stays the same, except that the internal search when processing a new submission will be more strict

these are the kind of graphs I'm working with - https://www.dropbox.com/s/puv4xx6fxne4fal/Scree... (the world is all one song, this won't get merged), https://www.dropbox.com/s/6534a8imxb0njca/Scree... (okay match, will be merged)

on the second graph, the blue line is an error value for the match, I guess?

it's the difference between two levels of "blur" (the blue and green lines in the plot above), it's normally used in CV for blob detection

and it's smoother out a bit, to avoid noise around 0

smoothed

the reason I'm doing this is so that I can tell when two fingerprint start matching, so I can locate a short sample in a long one and stuff like this

telling those songs apart is just an added bonus :)

cool

using my awesome application #39, "Kepstin is using curl to manually do lookups on the command line"

Is that available on github? Or do you run it as a service? :-P

it's currently version -2, which I think means you have to use a time machine to obtain a distributed copy.

reosarevok: i agree

not sure how that could work.. but I have seen some weird values at some points, and I assumed it was because something went wrong there, rather than the algorithm being totally fooled.. (I don't have the example handy, hence my asking around a little bit vaguely)

Gilou: No.

Gilou: For AcoustID, you can log in on acoustid.org and de-associate wrong matches.