Sound for Video Session: Loudness Normalize your Video Sound in Adobe Audition
ฝัง
- เผยแพร่เมื่อ 18 ม.ค. 2019
- How do you get loud audio for your video projects? What if the music is much louder than the dialogue or speaking parts? In this session, we cover the basics of mixing music and dialogue and show how to loudness normalize your video’s sound using Adobe Audition.
Please consider my sound for video classes available over at school.learnlightandsound.com
Gear used to record this episode:
AKG C414 XLS large diaphragm condenser microphone - this is my new, main voice over microphone
Amazon: geni.us/DoO4qT
B&H: bhpho.to/2ry8mmz
dbx 286s Preamplifier and channel strip - new vocal processing channel strip. I’ve had a ton of requests for help setting this up so I had to add one to my collection so I can learn how to use it first.
Amazon: geni.us/zrs0
B&H: bhpho.to/2CvIKfd
Universal Audio Apollo x6 - Thunderbolt 3 audio interface used to take audio signal from the dbx286S channel strip, convert it to digital, and send to my computer for recording.
B&H: bhpho.to/2CwRx3b
Copyright 2019 by Curtis Judd
Outro music from Artlist: Sunscape by Oliver Michael. You can receive a 2 week discount on a subscription to Artlist, a subscription service for stock music you can use with your videos. Check it out at artlist.io/artlist-70446/?art...
Ethics statement: Some of the links above are Amazon.com, B&H Photo, or other affiliate links.
Great explanation of loudness normalization, and a new look at waveform peaks! This is an eye opening tutorial, so thanks for sharing this knowledge!
Thank you, Dave.
The Mastery of Explanation - You Nailed it!
Thanks BFM!
Very in-sighting and helpful. Thank you for always inspiring me.
Thanks Franklin!
Thank you so much for making this simple!
👍
Always helpful - thanks!
👍
Super helpful! Thank you!
Thanks Daniel!
Thank you for your super valuable free knowledge, u r now my best tutor :)
Love from India
Thanks Fuddu!
Excellent info top to bottom. If only all TV networks would loudness normalize program material and ads then I could put down the remote and stop constantly adjusting the volume.
Agreed! Who’s enforcing that act congress passed about 5 years ago now?
Excellent video.
Thanks!
Great informative video!!
Thanks, Captain!
Yup sooooooo good!
Thanks!
Just watched this demo. Thank you Curtis. I'm 20 year veteran pro photographer and newbie to audio production. Now if I can only duplicate this in Davinci Resolve 16.
Hi Mark, thanks. I do have a DaVinci Resolve/Fairlight course which covers loudness normalization in version 16 if you are interested. You'll find the course over at school.learnlightandsound.com/p/fairlight-fundamentals-1
@@curtisjudd Terrific, Curtis! I have your course on the MixPre (I own the MixPre3 II) and using it with a Nikon Z6. Learning curve is straight up vertical and I'm attempting to augment my photography business with video skills. Your tutorials and the Learn Light and Sound site is tremendously valuable, well produced and professional. Thanks!
@@markholloway3208 Thanks so much, Mark!
I'm using auto software without LUFS measure, so I stuck with the "ears method" You mentioned. But it's good to know the basic theory.
👍🏻
Thanks for the video Curtis. Question: if I'm trying to get a TH-cam video to be the standard (-23LUFS?) across the entire video, what would be the best way to get that done in a timely fashion when editing in premiere? Do I just link all of the audio to audition and drag each clip into the match loudness window? How would that affect all of the audio effects I've applied in premiere including mixer effects that go across an entire track? Thanks in advance
Hi, you can do it that way or just use your ears to get each of the clips into the same ballpark and then use the loudness normalization export feature to do the final loudness work on export: th-cam.com/video/RrRn8stkEBs/w-d-xo.html
What if tracks are 2xmono (podcast voices) and one stereo (music) - how to approach match loudness?
Great content
I’m having a band practice session recording in both video and audio coming straight from mixer outputs
I’m taking the multitrack recording home for editing and overdubs, how can I put back the edited audio file in sync with old video
Thanks Jorge. All of the major video editing apps have a feature to sync audio to video clips and they line things up by matching the waveform of the video clip to the audio clip. Works really well.
Great tutorial Curtis! Just wondering, is there any way to do loudness normalization in DaVinci Resolve?
Hi Antti, yes, there is a loudness meter and you set the target loudness level in the settings. The process is a lot more tedious in Fairlight because it does not have the offline Match Loudness function and does not have an offline Loudness meter, just a real-time meter. We cover how to do this in my Fairlight mini course over at school.learnlightandsound.com
Do we need to compress audio before match loudness or match loudness does compression? Anyway it's very interesting what effects the match loudness applies to audio.
Yes, in most cases I’ve found I need to do at least a little compression.
1:25 Start | 2:00 Get both in the same ballpark | 2:20 Take source tracks in Files panel and use Match Loudness panel by pulling both Files and dragging them onto the ML panel | 2:30 Set both for -23LUFS 1-dBTP | 2:40 Run Match loudness | 3:35 Compress VO | 3:40 Open Amplitude statistics panel and Scan | 4:42 the math part | 5:05 Add compression | 5:47 Scan Amplitude Statistics again | 5:59 Back to multi track | 6:42 Mixdown session | 7:04 Match loundness / LUFS explained | 7:30 Rise mixdown to -16LUFS | 7:55 Amplitude stats of mixdown verified.
Thanks!
Great tutorial, you have a knack for explaining things! Quick question - if I understand correctly, we compressed the audio (5:11) because we needed more headroom in order to increase the loudness from -23 LUFS to -16. But we're the ones who cranked the loudness down to -23 when we initially matched loudness (2:29). Couldn't we have set that a bit higher, say -20, and given ourselves enough headroom right off the bat so we could skip the compression part and go straight to mixing down the track?
Hi Glyph Videography, great question. I only normalized in the first step to make the waveform easy to view for the tutorial. In either case, you'd still have to do some compression to loudness normalize at the end.
@@curtisjudd Got it, thanks!
Hi Curtis, great tutorial, does this LUFS option exist with Davinci Resolve and Fairlight?
Yes! You'll find it in the Fairlight panel.
another excellent video from Da Man. I would do it this way, but i'm not going down the Adobe route. I run FCP and Logic Pro, but it's a shame that there is no real competitor to Audition in the Mac world (at least that i have found). Have looked at Audacity, which is fine as far as it goes, but not really up to the kind of job. I CAN get this done in Logic, but it's not ideal for this kind of thing. But at least it plays ball pretty well with FCP
Yes, I’m still tempted to try Logic out...
Hi Curtis, now that you mentioned it in this demonstration, what is the order of operations when it comes to sound?
Hi MONIZ, good question! We covered order of operations here: th-cam.com/video/19jp1NgRZBw/w-d-xo.html
Would you please do a workflow similar to this on Izotrope RX 6 and the Acusnous plugins? TIA if possible ... Audition is just too darned expensive for us amateurs!
Hi Martin, I did this piece on RX some time ago: th-cam.com/video/B5Ll3cc1B7M/w-d-xo.html Note, however, that RX is not a mixing app so not everything applies. One still generally needs a mixing app, even if it is just a video editing app (NLE).
Thanks again! Great content! You're a wonderful educator! Is it possible to measure loudness within Premiere and whithout
the round trip to audition ?
Hi Michael, Thanks! I do not know of a way to measure loudness with a single click in Premiere. You can open a loudness radar plugin and play through the entire video to get a measurement but that is time consuming.
@@curtisjudd hi Curtis thx a lot. I know, that there should be a very hidden one deep deep down somewhere in the audio panel?! I saw somebody (a freelanc Premiere pro trainer) demonstrating it on a adobe roadshow. Anyway thx a lot for your great and helpful content!
Loudness radar, is my guess.
This was phenomenally helpful - normalization standards have always been confusing to me. Question: where does the -16lufs web standard come from?
Thanks Oliver. The -16 LUFS is really just a recommendation made by a group of respected post audio engineers in a white paper. I find that -16 to -17 LUFS works pretty well. Quieter than that and it isn't good for an audience listening on mobile devices in noisy locations. Louder than that and you squeeze so much dynamic range out of spoken word content that it sounds rather odd.e
Curtis Judd thanks man- very helpful info!!
This is fantastic, thanks Curtis. Is it possible to do this type of processes in Resolve (Fairlight)?
Hi Carlos, yes but it does not have a match loudness tool like Audition does, so it would need to be done manually.
@@LearnLightAndSoundSessions Will you make examples in Fairlight in future videos? I hope so. You are amazing! Thanks again.
@@carlos.o.flores I hope so, but first I need to find some plugins that will work with Fairlight to do at least the loudness measurements offline rather than having to run through the entire video piece in real time. It is very time consuming.
Hi curtis just to ask the following. I recorded my church band I did the video and someone else did the audio. The audio has been mixed with the vocals and music already and come as one track. With that being said do I still need to normalise the combined vocals and music as there is no way to separate the two. Any advise is appreciated. Thanks as always JF
Loudness normalization is just something I do as a last step on the final mix. So having it already mixed just makes your job easier. 😀
@@curtisjudd thank you curtis truly appreciated always enjoy you channel. You have a great day. J F
ok, million questions here. First: what were the steps before we got here? In Premiere I have imported and proxied the footage and synced up the audio. There is some wind noise, recorded levels are not perfect. So.. do make the whole video edit and only THEN move to audio at the very end in Audition via Dynamic Link?
(I will have to watch this about 10x to fully comprehend and internalise this solid flow of knowledge)
Usually yes, do the edit, send it over to Audition via dynamic link, clean up issues, then adjust levels so that they match, loudness normalize the entire mix, then back to Premiere.
What if i record Ambient sounds and there ar many quiet parts for longer duration but also very loud parts. If i bring the Loudness to -16Lufs the loud Parts need to be ultra compressed or hard limited and loosing all it´s dynamic Range.
So it would be better to lower the overall Loudness to keep the dynamic Range. But what will TH-cam do with that? Does TH-cam normalize to quiet Parts to -16Lufs anyway or does TH-cam only lower too Loud Parts and leaf the Quiet Parts as they are?
TH-cam does not boost levels to achieve a loudness target, only lowers them if the overall integrated loudness is higher than -14 LUFS. For ambient recordings, I would keep the dynamic range and NOT worry about hitting a LUFS target. It seems to me that LUFS targets are made with dialogue and music in mind.
@@curtisjudd agreed on that. Thanks for your Thoughts
Hi Curtis, you said that -16dB LUFS is optimal for web, yet you usually upload with -23dB. Is it because you assume that your audience has above average audio gear? And/or because audio is (naturally) important in your videos, so you don't want to compress it unnecessarily?
Btw, I use -18dB LUFS for lectures and interviews on youtube, it's a good middle ground, right? And I set 12ms look-ahead and 200ms release because that's what I found recommended for podcasts which is IMO quite similar to lectures and talking head style videos that I usually do.
Hi, for my main channel, which has videos that are more “produced”, I generally aim for -16 to -17 LUFS. This channel was originally meant to be live-streams without any editing. I knew I could commit enough time to record, but not necessarily time to edit and fine tune and properly mix so that’s the main reason that I don’t usually loudness normalize the videos on this channel. And, happily, most viewers have better audio playback systems as you suggested.
I struggle with the -16 target to be honest. Compressing to hit that target can leave some voices sounding rather odd. I agree that -18 is a decent compromise, but it can leave some of the audience straining to hear if they’re playing it back in less than ideal circumstances. Since you are aware of that risk, it seems like a reasonable compromise to me.
@@LearnLightAndSoundSessions Thanks for the reply! Have a nice day! :)
Many youtubers need this lesson. I hate having to ride the volume control.
👍
Hi Curtis, where can I find information on what target LUFS I should use on different usage, such as TV, web, Netflix, Radio, Cinema ads etc?
-24 LUFS - ATSC A85 (US TV), NPRSS & PRX
-23 LUFS - EU R128 broadcast
-19 LUFS - mono online, including mono podcasts
-16 LUFS - stereo online and stereo podcast
-14 LUFS - Amazon Echo/Alexa and Spotify
The first two are law. The rest are recommendations or what the companies listed just do regardless of what anyone else thinks.
@@curtisjudd thanks! What about cinema comersials, and cinema movies (feature and shorts) in 5.1 or 7.1 surround?
Unfortunately there aren’t any ratified standards for cinema. But I would go with -24 LUFS unless otherwise required.
How do you do this with a sequence that has multiple clips of dialogue?
You first mix down the entire sequence, then loudness normalize the exported mix.
I recently installed DaVinci Resolve to try to learn the basics of editing videos in that system and seeing what Fairlight can do for sound. The tools you show in Audition seem to be perfect for getting levels measured and set to what is needed for various media, web, broadcast, etc. Does DR and Fairlight have those same capabilities as Audition? I think the answer is no. If it were yes then I would likely buy your Fairlight class.
Hi Allen, no, Resolve/Fairlight does not have an offline waveform editor like Audition. However, I round trip the final mix down to RX where I do all of that work. It is nearly as fast as working in Audition - takes maybe 5 seconds to transfer the mix over to RX, when done there I click save and when I go back to Resolve, the processed audio is now a new layer on the same track. It actually works quite nicely. I'm nearly convinced to give up my Adobe Audition subscription.
Curtis Judd Audio It seems to me the best thing about Premiere is Audition! So to do what you do in Audition while editing in DaVinci I need to buy RX7? Looks like DaVinci is not free after all 😀.
@@RallenMan I relied on RX7 even when I did my mixing in Audition. So now, I find Resolve/Fairlight and RX to be a usable combination. I could drop Audition altogether, but I pre-paid for a year and there is one thing I have not yet figured out how to do with Resolve/Fairlight: When I'm editing/mixing just audio (no video), I have not found a reliable way to export the final mix other than to round trip it over to RX and export from there. There is an "export audio only" option in the deliver page of Resolve, but most of the time when I attempt to use that, the "Move to Render Queue" button is disabled and I haven't yet been able to figure out why. I thought I found the issue - that it wouldn't work if any tracks were muted - but it turns out that isn't the only factor. I just need to invest the time to work with Blackmagic support to work through the issue.
Hi Curtis, I know you recommend -16 for the web but everything I've read on TH-cam suggest -13. Which should I use?
Hi MONIZ, for spoken word content, I usually target -16 to -17 LUFS. Some of the music services target -14 and that can be ok for popular music styles but it crushes dialogue. So if most of your piece is dialogue, I'd targe -16.
So if it's a 4minute film with just music soundtrack then target -13, correct?
@@monizfilms Personally, I would always target -16 LUFS. I find the louder -14 or -13 too compressed and don't like the sound. But its up to you.
Thanks Curtis. You're Number One.
Should you EQ before you use match loudness?
If you need to EQ, yes. Loudness is the very last step.
Very useful!I have been fortunate that what I have edited offline has been dealt with at a MA studio during the online edit. My recent work however has not enjoyed the same luxury and both offline and online are now increasingly down to me.
I don't use Adobe so if anyone knows of a good tutorial that uses Logic Pro X to do the same thing, I'd be grateful if you'd post the link (I'm searching too of course, but what you know may be better than what I find). The techniques are obviously the same, but the applications are different. Actually, I wish Apple would up their game in terms of FCPX to Logic Pro X and make it easier and more intuitive.
Thanks Andrew. I don’t know of any specific Logic Pro X tutorials, but will let you know if I encounter any.
@@LearnLightAndSoundSessions Thank you Curtis. That would be great. Perhaps I should simply start using Audition, although I don't like 're-investing' in tools if I can help it.
@@andrewjones-productions If it is any consolation, I recently cancelled my subscription to all of the other adobe apps except Audition. I'm going to start working in Resolve (Fairlight) and might try Logic as well, but I suspect they're both missing some of the tools in Audition that I use quite often (e.g., ability to split up a poly wave to separate tracks, convert a mono track to dual mono, and offline tools like waveform statistics and offline plugin processing so that I can round trip to RX Audio Editor).
@@curtisjudd Yes, as I just posted above(!), Logic does have many useful tools, but not as specific as there are in Audition, the way you use it. I am not sure that Logic would be ideal for round tripping to Resolve, although I guess you could do it. It would really only make sense if you were editing in FCP as well, as the Audio roles functions in FCP will slot into Logic pretty well, Other editors, well, i'm guessing not so much. Might be worth looking at some of the older Ripple videos, as i think they have done some stuff on FCP->Logic which might help. Will investigate and get back to you.
OK... hope you don't mind me posting links to other YT channels, but here is one potentially useful video for FCP=> Logic. And remember, this id 4 years old, and both Logic and FCP have had many updates since then...(like proper colour correction wheels, etc. etc.!)
th-cam.com/video/dmOu8dfRA-I/w-d-xo.html
Perfect! notice a lot of youtube videos have their music way too loud, even music maker channels :\
Yes, most music is mastered to -10 to -14 LUFS. They sacrifice a lot of dynamic range when they crush it that hard.
@@curtisjudd Yea, I produce and mix songs and the goal is to get around -14 LUFS for commercial music. I'm new to video editing softwares and It's interesting to see other industries using LUFS to measure perceived loudness. Makes sense!
7:57 -1 -1,5
Context, I assume Voodu is referring to the target dB True Peak level.
Here is a podcast I host and based on this video I have be using these skills and getting better! anchor.fm/elderqueen/episodes/The-Church-is-a-Place-You-Go-with-Cameron-Shaffer---S2---Ep1-e3vm7s
Nice work!