Headphone calibration with just reverb?!

[@NexxuSix]

To simply say “Thank you!” Doesn’t do this video enough justice! I experimented with my headphone profile (Sennheiser HD 280 Pro) and could tell there was quite a difference when the Hybrid Reverb was turned on and off. This is truly helpful! Thank you!

[@LOLCoolJ]

This is very good and useful. I would add: If the user's choice of correction includes boosts rather than just cuts, it would be wise to decrease the overall volume level to compensate for whatever the highest boost may be. So if your highest boost is a low shelf of +3db, the over all level should be reduced by 3db. I hope you and your family are doing well. :)

[@SeedtoStage]

I meant to add this! Thanks for bringing this up.

[@scottmyers6564]

to monitor through headphone correction I make an audio effect rack last on the master, create two chains - one chain that goes straight to master out (hardware 1/2 for me) and one with headphone correction that then uses Ableton external audio effect to route the corrected chain out of a different hardware output (hardware 7/8 for me) with no "audio from" input. A little convoluted but only needs set up once and makes it so you don't have to worry about bypassing the correction before bouncing. awesome tip, going to try out this instead of EQ.

[@SeedtoStage]

Nice that’s a good call for folks who have multiple outputs on their interface

[@Grant82gc]

This is a brilliant idea!!

[@Mortadella555]

wow, worlds collide!! I can't say I'm not surprised that my favorite ableton youtuber is also a github user!! (I'm a software dev. and a beginner wub wub ableton learner.) you are salt of the earth, man.

[@MadNoisy]

nice pfp 🤘🤘

[@flyoverfredusa]

best music production channel bar none

[@sillyviep.1956]

I agree soooo much!!! 🤩

[@robbiep742]

I say it every time.

[@PhilRushworth]

100%

[@DavidAsh42]

Thank you, Anthony. Blessings to you and the family.

[@Bthelick]

regardless of frequency content, the issue is still that headphones aren't stereo. You still need to add back in some interaural time differences to get near the benefits of speaker judgements.

[@kishtamusic]

u can use a mid side delay and delay the mid by a few ms or even less and that gives you that feeling like the left is a bit on the right and vise versa

[@Bthelick]

@kishtamusic it gives you the feeling yes, but vitally still the right ear can't hear any of the left and vice versa unless you introduce cross talk, (delayed preferably)

[@moosey62]

Great presentation mate, thanks a lot. I'll be on this tonight.

[@deardamo]

Thanks man, this was super helpful!!

[@54LZ]

I think this is even superior to using an non-linear parametric EQ because the convolution reverb is not altering the phases of the signal and thus only "shapes" the original signal with the headphone harman curve.

[@SeedtoStage]

See that I’m not sure of, I don’t know enough about DSP programming to know BUT if you were to record music played out of speakers and try to null test it against signal with the exact same frequency response, it still wouldn’t null because of room reverberation. So I wouldn’t sweat it too much.

[@thomaslthomas1506]

After a bit of hair pulling got this to fly on my 240's Which are all the headphones in my room. I have 2 outputs on my interface that are stereo mixes for phones. So I set up a sub out in Ableton and sent to each of the headphone feeds and bypassed the master to headphones. Viola no forgetting to turn on plugin for mixdown. Then made a template for all new files being created in Ableton pretty slick. Only works if you have at least 2 discreet stereo outs on your interface though.

[@himothyjah]

Thanks so much man, this is awesome!

[@WillCarlotto]

Awesome video. I don't make beats nor use Ableton, but your video was so useful. And I'm looking foward for the "free solution for crosstalk issues" that you mentioned.

[@WindMakesWaves]

Awesome video! I recently heard about the Harman Curve and have been meaning to learn more! Thanks!

[@themorphingorb]

Very nice walkthrough! I finally decided to dive down into this after seeing that SoundID charges 100 dollhers for their service! I used the 'convolver' from Kilohearts to load in my impulse response. It's nice and sleak and doesn't have any bells and whistles; ergo, it's hard to f up xD haha

[@Horizontedesucesos_]

Your channel is the best, man. I'm going to try this tonight (if I can find my headphones there). 🙌🏻

[@ChrisG-n6f]

Love your videos man. This ones awesome, great explanations.

[@josipvulic5471]

Love this! Such a great method to only use headphones for those with small apartments! You are a rabbit!

[@johnviera3884]

Logic Pro has a stock convolution reverb plugin called Space Designer.

[@gothboisinner1935]

i saw fl edison plugin and it makes blur effect how can i make similar effect in ableton ?

[@lzcim]

Tip: use soundsource (mac) dunno if them have win version, so you can open any vst and do the calibration outside your daw, you will not have the problem of have to turn off when exportin and the calibration will effect every sound played in your pc, witch I think's the best cuz you have to adapt to this curve. Also use "Ghz can open 3: Mix enginer" preset at the end to help w/ "interaural time differences".

[@mouderni]

Never thought of this. Thanks for the tip @lonczinski . When adding an effect (in soundsource), I saw the Headphone EQ option, and was surprised to see the measurements from GitHub. Played some music and could instantly compare my headphones multiple profiles. Also went with one from oratory1990.

[@nicoincertezza5763]

hey teach, great useful information once again. thank you!

[@thesageofvisonaryroots]

Thank you so much for this lesson!

[@incantic]

Awesome. This makes perfect sense as an eq curve and an impulse are the exact same thing just transformed differently mathematically. I have my eq curve in sound source. I also use a plugin called can opener to do the cross talk which helps amazingly with mixing stereo in the lower frequencies. I like the sound source + can opener approach because it works outside of my daw so I never have any chance of exporting a project with an eq curve on it. I lost the link but you can actually find crosstalk impulses as well and add that to your chain.

[@incantic]

kzbin.info/www/bejne/qGiYZmmvl9mXjZI This video on eq is rad btw

[@robbret]

If you put this on your master and you export your track to audio through your master, you'll have an incorrect eq balance, right ? How do you get around this, (except for having to remember to turn it off before exporting) ?

[@nsjx]

Stellar tip. I gotta trycthis and compare it to my existing Sonarworks Ref4 calibrated hd650s. Been using this setup for over a decade with CanOpener for crosstalk. GoodHertz has some great articles on the x-talk topic. What you say in the end is important to remember. Onw can do just fine if one knows the usual deficiencies of one's room or monitors. It really helps to analyze similar prof produced music to see how stuff translates in your own equipment. Anyway, thanks for the great thoroughly explained tip. I learned something.

[@nicholasbelvis732]

I don't understand why you would bypass when you export it? Unless you'er printing you're mix on a separate audio track. (in that case it makes sense cause that you'd be applying the impulse twice) If you are making all of your decisions mixing into this impulse and then bypassing you're going to be hearing two different tonally balanced results? I did this my self and after a bypassed export, its as if all of the precise adjustments i did were completely misrepresenting what was actually there tonally. Im mixing on AUDEZE M500s (Also tested the results on my Genelec 341As with and without the GLM calibration) Bypassing the impulse typically left the mix muddy and re-introduced a hand full of resonance issues in the mids that were not audible with the impulse active. Would love to get some more insight into this, but how I see it getting your ears to learn your headphones and how they translate onto other systems is key. Love all the content!

[@SeedtoStage]

Think of it this way. The goal of this curve / calibration is to alter the frequency response of your headphones to emulate the sound of speakers in a room to help you make better decisions and listen at an appropriate level. If you export the music with the curve applied, the listener A, does not have your same equipment, and B may be listening in speakers. You’d be baking in a correction meant only for the headphones you use.

[@foodforthegods]

This is sick! I tried the Sonarworks thing ages ago but I wasn't a fan of the software and how it's treated as audio "hardware" by Windows. This is much better. Kinda crazy how much more pleasant it is to listen to. The mid boost sounds a bit off at first but it's great after the ears have adjusted. Also noticing how much louder I can listen without feeling overwhelmed. Huge bonus too, I still use foobar to listen to my music library on PC and actually found an IR convolver component!

[@iambadd]

It's annoying though that you can't listen to music normally with the same EQ adjustment as the reverb result. You'd go from no correction to corrected EQ and all of the stuff you just listened for reference is sounding different

[@foodforthegods]

@@iambadd You could get a file of the music you're referencing and put it on its own track in your mixing project.

[@danielweirdone]

this is fantastic... thanks!

[@animatedgifmusic]

This is super handy! Thanks for sharing

[@TUFAtunes]

This is dope. How you like those Ananda plannars? I'm looking for something hifi I can dial in a little better than my ATH-40x, something with a little more sub range as a produce bass tunes. They seem decent for the low low end?

[@SeedtoStage]

Not only better low end on these Anandas, but can handle lots of SPL in the low end when cranked with the harman curve. So GREAT for making bass decisions. But still prefer my speakers when I can actually use them.

[@NTidelius]

so, so helpful. Thank you!

[@aidennymes6335]

sonarworks hates this trick

[@MadelnMachines]

I notice the convolution seems to get closer to the harman target than setting EQ in most cases. Does using convolution also have more or less potential phasing issues than Ableton EQ. Do you need to match the 44khz or 48khz IR to your Ableton settings?

[@SeedtoStage]

Whatever phasing it would cause would likely be much less than a small apartment room without much treatment is how I see it. As for the sample rate, I would choose the one associated with the same rate you use.

[@MadelnMachines]

@@SeedtoStage Cool thanks

[@TheHutchMusic]

I love this channel... Thank you.

[@KristijanJankovic]

holy smokes what a cool video and thanks for the great tip

[@gnomescape]

Wow thank you. Seen other videos on the ?Harmon curve and been scepticle as they always talk about boosting bass and I have othe rproblem. I hear bass when I mix but it vanishes on youtube. But as this involved impulse responses and I have KHS Convolver I thought give it a go. When I looked at the graph it showed my Beyerdynamic DT 240 pro headphones do indeed have too much bass and needs reduction. So I got exited. Downloaded responses and plugged them in... Amazing. I can make the mixes the sounded amazing at home sound as bad as they did on KZbin!

[@yvancarbuccia6596]

what if i cant find my headphones in the database?

[@kalambalam]

Dude, I thought you were in Nasheville. I had to leave Asheville too. So sad!

[@e_abstraction]

Didn't realize they had wav files on that GitHub I just made an EQ curve initially. Do you know if using the convolution has some delay compared to an EQ? That would be the only thing I'm worried about.

[@SeedtoStage]

I’m not an audio scientist but my guess would be that because the IR is only a millisecond or 2 long, that it wouldn’t impart any real delay to the signal. Hybrid reverb creates zero latency.

[@e_abstraction]

@SeedtoStage awesome I just implemented it into my writing template 😄

[@SpeedRacerx89]

Damn. I have Ableton standard edition and it doesn’t have the Hybrid Reverb tool.

[@SoundsFromGerome]

that is super helpful

[@hirnzircus]

Great, thank you.

[@bobbyrinehart1299]

You are a genius! 🤘

[@harryteunissen5332]

Cool, but what I miss in all those discussions is: what happens inside your head? How good or ' calibrated' are my ears? I know in my case I need a hearing aid for mainly the frequencies from 7,5K and above. So, how do I translate that áfter the calibration of my headphones? I think I will need another EQ to compensate for my hearing loss, and so will many producers do. The main question we need to answer is: what do I actually hear if I compare it to others? How to find out? With a sort of threshold EQ/Gain/Pink noise level?

[@beatskool101]

I tried the Harmon Curve after the master limiter and found everything way too loud and clipping , because it was no longer -1, then I realized my eq needed the pre-amp to be -11.2db, so I changed it(with the gain in the eq, as it was the only place to do it), and now obviously everything was too quiet, because it's no longer the -1 that the limiter is set to, so even if I went in and altered everything, I will never hear the difference, what am I missing?

[@SeedtoStage]

@@beatskool101 your audience who hears your mix is never intended to listen to the Harman EQ. So if I was you, I would simply place utility between your limiter and harman EQ and turn it down to avoid clipping. Once your mix is ready for export, simply disable the utility and harman.

[@beatskool101]

@@SeedtoStage I know not to export with it but it makes what the master compressor & limiter is producing 11 db quieter than it is, so if I adjusted levels based on this, I dont think it would be correct, even though I know some of my things probably need to be quieter.

[@thomas-vk4vc]

Why even use sonarworks anymore?? This is amazing!

[@SeedtoStage]

well sonarworks has their own target curve that they use which is slightly different than harman. They also have linear phase eq and some other features. I do think its odd that they dont offer just a straight up harman curve in their options. Maybe they would have to buy a license for it or something. Who knows.

[@DiMK_46]

I just bought the KRK Rokit 5 that my beloved Dj Mandragora uses (Rokit 7) and my room is almost untreated (only acoustic foams). Did I buy those pointlessly ? I ve seen big difference compare to the M Audio BX5 that i had before.

[@SeedtoStage]

@@DiMK_46 of course not! You can mix on anything in any environment if you learn your system and your own tendencies. If I was you I would get a pair of open back headphones and check your low end on the headphones. Also I would beware of foam. It only attenuates high frequencies leaving the low frequencies untouched meaning you’ll tend to mix brighter than you might mean to due to the treble absorption.

[@DiMK_46]

@@SeedtoStage Yes you felt in! Exactly! For the moment my mixes tend to sound brighter and then i return back to ableton trying to fix it. Thank you so so much💯💥

[@aidennymes6335]

untreated rooms always have "dead" frequencies at diferent listening spots. no matter what monitors or correction software you use, you will never be able to hear those frequencies. so you are better off investing in propper treatment (bass traps and akoustic absorbers) and/or good pair of headphones + correction (i have dt990 pro open back for better stereo and m50x closed back for better low end). and dont forget to add crappy 30€ pc speakers to check your mix on them as well ;)

[@DiMK_46]

@@aidennymes6335 Hi friend, thank you so so much for your advice and precious information. But how Dj Mandragora produces in his kitchen.... no treatment at all.. how how how.. damn.

[@aidennymes6335]

@DiMK_46 i guess someone else is doing mixing and mastering for him in a propper studio.

[@aesonone]

there is only the itsy bitsy issue remaining of the measurement itself being done on a particular simulation head that is different to yours, so you cannot really get a perfect translation, compounded with the fact that the harman target is aiming for a population average, that you personally might happen to lie pretty far off of anatomically. still a pretty neat trick for chucking a calibration target into your daw tho

[@SeedtoStage]

Yep this is 100% true and why there are so many different opinions about it. I will say this though, in most cases it will be an improvement over most consumer cans. The main take for me is that the midrange area of EQ curves seems to be agreed upon among many of the different targets and their only significant deviations are in the treble and sometimes the low end.

[@nuladvajedan]

Hey, I have RedDragon Ares H120 and shame on me those are some shitty gaming headphones. My question is how can I make them sound better because they are not listed on that github page🤔

[@sebguyader]

I discovered that recently, after watching Paul Third's video. And indeed, using the IR and convolution gives better results than parametric EQ.

[@MrMrMrMrT]

This would be so awesome if they had Munson curves as well. Then you can a/b/c Edit. Actually I had it backwards. The curve you’re talking about isn’t meant to be a flat curve. It’s supposed to be a curve that sounds good eg some thing that will translate and be what the end user hears

[@LukeIcardMusic]

I subscribed when I heard the word NC. Bro im in nc too! 😊

[@omnione7894]

❤❤❤❤

[@burn_after_reading]

How on earth does this work? Does the white noise disturb on the background while mixing. I'm sure not, since it's needed for the headphone calibration, but it's hard to wrap my head around.

[@SeedtoStage]

The white noise is just a proof of the effect. You don’t use white noise in the actual process. Instead you put the convolution reverb with the headphone profile on your master / main track and listen to you mix through it. Then you disable it and export when your mix is done.

[@burn_after_reading]

@ Thanks for your quick reply! Will be nice to test and hear the technique in practise tomorrow :) It’s now night in Finland.

[@SoundsFromGerome]

😎😎😎

[@kishtamusic]

well i just tried 3 different files for my beyer880 dt and they each sounds very different from the other. all 48khz minimum phase haha.... nobody to trust these days lol

[@SteamvilleQuintet]

Ah, but what is the Harman curve of the little celluloid phone speaker 80% of our audience is listening too? Shouldn't that be "factored in"? How many listeners out there have flat field studio monitors, or do we master to impress other studios?

[@SeedtoStage]

The point is to have a listening EQ curve that is playing back as true as possible to the experience of a flat response. That way, crappier systems that deviate from that good mix, won’t deviate as much as they would if the mix was already bad. Hope that makes sense.

[@SteamvilleQuintet]

@@SeedtoStage I'm reminded of the days when there was a "Not For Sale - Promo Use Only" version for radio stations listened to in cars and cheap portable radios; then the real mix in the stores for the Marantz people. But if some folks are accurate, the world will soon be 7-1, 9-1, 16-1 surround encoded, which will make a lot of VSTs and methods pushed at us now useless.