Matchering 2.0

All your LV2 and LADSPA goodness and more.

Moderators: MattKingUSA, khz

Post Reply
User avatar
milkii
Established Member
Posts: 477
Joined: Tue Jan 05, 2016 9:08 am
Location: Edinburgh
Has thanked: 92 times
Been thanked: 91 times
Contact:

Matchering 2.0

Post by milkii »

Matchering 2.0 is a novel Containerized Web Application and Python Library for audio matching and mastering. It follows a simple idea - you take TWO audio files and feed them into Matchering: TARGET (the track you want to master, you want it to sound like the reference), and REFERENCE (another track, like some kind of "wet" popular song, you want your target to sound like it). Our algorithm matches both of these tracks and provides you the mastered TARGET track with the same RMS, FR, peak amplitude and stereo width as the REFERENCE track has.

* https://github.com/sergree/matchering
* YouTube: Matchering 2.0 - Open Source Audio Matching and Mastering
* YouTube: Matchering 2.0 - How (not) to Use It

they/them ta / libreav.org / wiki.thingsandstuff.org/Audio and related pages / gh

User avatar
funkmuscle
Established Member
Posts: 2800
Joined: Mon Jun 02, 2008 2:30 pm
Has thanked: 129 times
Been thanked: 31 times

Re: Matchering 2.0

Post by funkmuscle »

Is this like Landr? Tramp also made a matching EQ.
User avatar
Loki Harfagr
Established Member
Posts: 268
Joined: Thu Aug 02, 2018 1:28 pm
Has thanked: 151 times
Been thanked: 53 times

Re: Matchering 2.0

Post by Loki Harfagr »

Hey, that's a nice full of promise one! Thanks for the note :)

Just been testing some mix-mastr-blund-unblund and cross-reverse simile and it's a very promising start!
Only some minor 'curiosities' in the overall limiting when using quite extreme refs but that was quite close to the usual matter of don't request your master to sound like Metulucuh when your orig stuff to mixmast is Graemenobad

And in any case that'll give some help and scents on where to start from for the fresh noobies around (as opposed to old noobz as me or I, indeed)
User avatar
logicgate
Established Member
Posts: 37
Joined: Sat Oct 20, 2012 2:21 am
Has thanked: 1 time

Re: Matchering 2.0

Post by logicgate »

I installed the command line version of Matchering. It seems to work pretty well after I figured out what options I wanted.

To preserve dynamic range, I use the --no_limiter option and avoid the --dont_normalize option. The --dont_normalize option works opposite of what I expected.

Thanks to the Matchering team for making this available as free software!
Logic Gate Music (retro electronic music and more...)
User avatar
GMaq
Established Member
Posts: 2774
Joined: Fri Sep 25, 2009 1:42 pm
Has thanked: 520 times
Been thanked: 555 times

Re: Matchering 2.0

Post by GMaq »

Hi!

This is very cool! To be honest I wasn't a docker user and kind of didn't want to be bothered installing something from yet another 3rd party repo but I'm glad I did! Great work!
User avatar
Capoeira
Established Member
Posts: 1321
Joined: Tue May 12, 2009 1:01 pm
Location: Brazil
Has thanked: 3 times
Been thanked: 2 times

Re: Matchering 2.0

Post by Capoeira »

Does this realy work? I imagine blindly copying EQs of a reference beeing complicated if instruments/voices don't match.
tavasti
Established Member
Posts: 2047
Joined: Tue Feb 16, 2016 6:56 am
Location: Kangasala, Finland
Has thanked: 369 times
Been thanked: 208 times
Contact:

Re: Matchering 2.0

Post by tavasti »

Capoeira wrote: Fri Jul 24, 2020 8:26 pm Does this realy work? I imagine blindly copying EQs of a reference beeing complicated if instruments/voices don't match.
Sure if you feed in Meshuggah and ask it sound like Dire Straits, results won't be optimal :-)

Linux veteran & Novice musician

Latest track: https://www.youtube.com/watch?v=ycVrgGtrBmM

User avatar
Capoeira
Established Member
Posts: 1321
Joined: Tue May 12, 2009 1:01 pm
Location: Brazil
Has thanked: 3 times
Been thanked: 2 times

Re: Matchering 2.0

Post by Capoeira »

tavasti wrote: Fri Jul 24, 2020 9:41 pm
Capoeira wrote: Fri Jul 24, 2020 8:26 pm Does this realy work? I imagine blindly copying EQs of a reference beeing complicated if instruments/voices don't match.
Sure if you feed in Meshuggah and ask it sound like Dire Straits, results won't be optimal :-)
well, you can construct a less dramatic example:
you are a Dire Straits cover band but you kick drum is tuned lower
Post Reply