i want to ask.. i record my video on smartphone and i have my record voice in -3 db but the volume on record is on 50-60% and i try to put music to see the bar its on -3 db but the volume its on 80 - 90%... so did i must gain up the volume my mic right? i confuse in here
Curtis, if I had wanted a mouthful of something, I'd go get a Hardee's #3 double burger... Good stuff, sir! This is one of the many things that plague newbs on the processing side of things, myself included. Thanks again!
@@curtisjudd I found the Loudness meters in my NLE. They are in LUFS, great. But there are 4 measurements: Momentary, Short, Integrated and Range. I guess that the numbers you gave for YT or Spotify is the integrated. But what about the Momentary and Short, what are good values for those?
@@dlapierre99 Yes, integrated is the measurement of the entire video but you have to play though the entire piece to get the final integrated measurement. The others are for analyzing and solving issues at particular spots within the piece, e.g., maybe I need to compress or level out this little bit here...
@@curtisjudd Thank you so much! I have an option to generate a loudness log during render so I can get the integrated value quite fast if I render only the audio. And I am able to graph each 4 metrics over time in Excel. Thanks a lot again, now I am able to understand what I am doing.
3:39 “It is an old maxim of mine that when you have excluded the impossible, whatever remains, however improbable, must be the truth.” ― Sir Arthur Conan Doyle, The Complete Short Stories of Sherlock Holmes As a Sherlockian, I got distracted by the quote! Hahaha...
This video is great! Thanks Curtis! Typically when I edit a short video that has loads of different sound pieces, I edit each one separately and do a loudness normalize to it. I'm not sure if it is the correct way or I should export the sound as a whole and then make the loudness normalize to it. Will you be touching it on the next video? Thanks again!!
Hey Curtis, here is a problem I want to solve with audio. My wife and I do UA-cam videos about RV life, We both use lav mics which go into our zoom h1 split into left and right Channels. I do this so I can control our audio separately, I end up having to cut out my Audio when she talks and vice versa. (There is an echo sound of picking up each other’s mic.). This takes a long time. Is there a way to do this automatically in premiere pro? Thanks Larry
Hi Larry, I'm not aware of an automatic way to do this in Premiere (but I haven't used Premiere in several years so perhaps they've added an auto mix feature?)
Thanks Curtis, but you mentioned something which doesn't immediately seem related to the topic here - you mention that vocals are *compressed* when normalized to the recommended loudness. It's not clear to me why amplification results in compression? Where does the compression come in?
Hi Richard, that will become clearer in the next video on this topic. Usually one will need to do some dynamics control (compression) to achieve a target which is louder than -23 LUFS for spoken word content. Compression is almost always part of that process or some of the audio transients (short peaks) will be heavily limited.
Yelp! Going back some time now, on one of your videos you recommended -23 LUFS (LKFS in Resolve) as the level to normalise your audio. I've been doing this for some time now to provide a consistent audio level to all of my videos. Should I now change that to -16?
It depends on where you publish or broadcast. For broadcast TV, they have specific requirements. In the EU, that is -23 LUFS. For the US, -24 LKFS. For online, I generally recommend -16 to -17 LUFS.
@epiclightmedia is that one channel that I had to unsubscribe from due to their excessive amount of helpful high-quality videos that helped me get better ;)
What I do is take live calls on live stream and have six dedicated wattsapp lines.. The audio technic used I have learnt through this your channel which I appreciate and THANXX u loads.. 👍
Fantastic video. I found your channel while looking for a microphone and you've been dedicatedly putting out seriously high quality content for what looks like more than a decade! Props!
I'd like to point out that television shows on TV have different technical requirements for audio than the commercials to. That allows TV commercials to be louder than program material.
@@curtisjudd I'll clarify. Television shows and movies are produced for release on their platforms. Movies may be adjusted before airing to adhere to LUFS/LKFS guidelines too but TV stations often compress any program content. They also visually process shows too which is why TV is a terrible place to watch a movie or TV show. These are done within guidelines (which are minimum specs) to make it a better experience on TV. Movies and TV shows are produced to look and sound a certain way and if it were possible to air without changing anything, it would be ideal for those of us that worked on it but there are broadcast and cable cast regulations that must be followed. I worked on a show that aired on USA and me and the sound mixer heard all kinds of distortion and compression when the show aired because that's what the station does to all program content. My mixer recorded the show and sent it to post and asked them about some of the things he was hearing. They responded back that their master didn't sound like that and this was common in TV. If we bought the DVD of the show, it would sound better. He was right. It's been over 2 decades since I worked at a TV station and that was pre-HD but I'm sure stations still have safety measures to make sure audio doesn't go out too loud or that dark scenes don't go too dark. Commercials, on the other hand, want to be noticed and may do tricks to get attention. Extreme compression is one truck they'll use so that if you're watching a late night show and doze, you'll wake up and watch. Another trick they'll use it playing the A-weighting curve - boosting the crap out of mid and higher frequencies because you hear them better. Really annoying infomercials may even apply aggressive low and high pass filters so that there's more headroom for an annoying presence boost (for example). Commercials don't have to be cinematic masterpieces, their goal is to sell a product so if that means using tricks to sound louder and be noticed, they will. Commercials aren't immune to processing but aren't afraid of playing games.
@@SoundSpeeds No need to be brief here, sir. Tell me more... Hahaha, thanks for the wealth of input. Your information coupled with what Curtis conveyed tells me that by sustaining certain loudness levels with very few quiet moments, infomercials and such can be perceived to be louder but still fly below the standards that mainly account for dynamic range differences. Am I understanding you both correctly?
@@insanejughead I think I understand that and I think that's right. Curtis said sustained loudness over time is perceived louder than audio with more dynamics and that's 100% right. What I added is that sometimes commercials and infomercials will use tricks to maintain loudness for the entire time and/or boost frequencies humans have a more difficult time ignoring while cutting those we may tune out.
Your content is amazing. Thank you so much for this. I've been struggling with audio, particularly when it comes to loudness for a while. Part of that is me, and the fact that I feel the need to news commentator "shout" on camera. But there is also the technical factor of where I should be in terms of audio signals. And this helps with that tremendously. Can't wait for the next video where we learn about tooling.
This has always been one of the most confusing things about video production. When I made my first TV commercial, they kept rejecting the video for two reasons... the loudness and the luminance / whiteness levels. The same side of two different coins. I wish I saw this before all that trial and error. I would love to have a button in Final Cut Pro that lets me just click a dropdown to use the correct levels for various platforms.
Sadly, I don't think we'll see that in FCPX. It doesn't have the concept of a master audio bus which is where the final loudness normalization would need to take place - unless they make some BIG changes to FCPX.
@@curtisjudd Without industry standards in FXPX the best we can do is use their generic Loudness Amount and Uniformity percentages which I set at 40 and 17 respectively for UA-cam. My newer flow is 32 bit float using Zoom F6 and MixPre 6 II, so I've been doing normalization in Reaper first.
Thanks Curtis, happy new year to you! Great video as always! Would you be able to recommend a sweet spot when working with something like very dynamic live orchestral for UA-cam? Thanks
I'd recommend trying different targets to see how they each sound and choose from there. Best if you can get it to -16 LUFS and be satisfied that it doesn't sound too compressed.
Curtis, that's amazing, I've just sent a question about LUFS on UA-cam in your online course, and you came with this video:) anyway, the question is - what is the right LUFS level for the UA-cam videos in 2021, for mono and stereo? Updated: I've reached the end of your video and finally have the answer, thank you! Funny fact from Russia: the "statistics for nerds" was translated by UA-cam as "statistics for system administrators ", so we're lacking some humor there🙃
@@curtisjudd And what do you recommend for videos with a mix of spoken word, music and sound design? I think you have to crush the dynamics too much to keep it on -14 LUFS without distortion.
You were reading my mind - I was recently wondering about those LUFS and experimenting to get a good level. So, I just changed your 444 likes into 445 ;-)
Glad to hear the remark at the end, it sounds like you're going to cover the question I was about to ask. I use FCPX with access to LogicProX, and want to know how to estimate my LUFS before I upload the video, ha ha. I always seem to undershoot, compared to professional style videos that somehow get it nearly perfect on stats for nerds.
Awesome video, thanks :) Just bought your "Dialogue Sound Post Processing with Adobe Audition" course, do you plan to record a course for Isotope tools? Looks like you're using Resolve instead of Audition for most of the mastering.
Still, I wough HIGHLY recommend adding this information to the course promo/landing page/description. So it's clear for potential buyers that the course is even bigger/better/etc. It would affect sales in a good way for sure.
Hitting audio standards in Final Cut is frustrating. Or else I need to learn the tools better. In any case, I appreciate it if you would address workflows outside your own Adobe ecosystem.
I totally agree. I finally got frustrated enough with FCP and bought a copy of Izotope RX8 (though there are many other capable tools as well) to normalize all audio before importing audio separate from video. RX8 has output presets for broadcast, streaming, podcast, and more. Adding this step takes a little longer at first, but knowing that my audio levels are correct for output has reduced my level of anxiety with the finished product. No affiliation with the company, just a satisfied user.
Hi Danny, I use FCPX as well. But I don't do the audio processing there for the very reason you cite - it isn't good at that, especially when it comes to hitting a loudness target/standard. I do most of my loudness processing in Izotope because it is fast.
I've only been following you for about a year, Curtis, but I've learned so much in that year. Thank you not only for providing your insights and knowledge but also for giving those insights "handles" and showing us how to use them.
So my question would be in Streamlabs what should be my dB target at any given time that would most closely correlate to - 14 LUFS, - 16 LUFS and lastly - 17 LUFS over time? I get that they aren't quite the same thing but I believe there should be a target dB level that if maintained most of the time that should correlate to a given LUFS, right? During my live streams I have game audio from a game like Call of Duty Cold War, music from Epidemic at 320 kbps and my voice from me speaking into a SM7B all mixed together in real time via my GoXLR mixer. So there usually is a pretty constant stream of audio sounds as one of the above is almost always playing, usually several or all even at once.
Is there a way to do Loudness Normalization in Premiere itself instead of taking the route through Wavelab/Audition? I end up using so many different softwares for my videos so I'm trying not to add more steps ;)
I have been doing youtube stuff for nearly eight years, I am better with my audio than most, because it's the first thing to bug me (and many, many others) in a video. But this was very helpfull, thank you!
@@curtisjudd Well, I have only you to blame for that. Last few months, you have made me deep dive into audio and get familiar with tools and technology I did not know existed.
@@TechforToastmasters Welcome to the nerd club, nerd. 😎😁 Now you're stuck in this rabbit hole with myself and the rest of us. We're a friendly lot, though...
@@insanejughead Indeed. When I first started, I thought buying the best microphone would solve all my problems so I went and bought a SM7b. That could not be further from the truth. Now my room is fitted with thick carpet, lots of soft furniture to dampen the sound, and paying close attention to levels, clipping, compression, equalizer, etc.
@@TechforToastmasters I don't have the SM7b, but I do have two Unidyne III 545's. The information I have found implies that the capsules are the same between the classic 545, SM57, and SM7b, except that the housing and transformer is different to create a subtly different sound for each. Actually, isn't the SM7b transformer-less? Otherwise it wouldn't need so much external preamp gain, right?
Curtis, you are the MAN. Do you have any idea how much I've learned from you, and how much those lessons have had real life changes in my workflow and understanding what is going on, so I can make better choices? Thank you a million times over.
The recommendation is -16 LUFS for YouYube with primarily voice. Got it. What happened to -23or -24 LUFS? Only for broadcast? BTW, saw the normalizing video. fantastic. Thanks!
-23 LUFS is the EBU target for broadcast television. -24 LKFS is the ITU target and is required for broadcast TV in the US. The internet is the Wild West with no requirements. My recommendation is -16 LUFS for online content.
Thank you so much for all your content and excellent delivery and production quality. Straight to the point (no BS) and textual summaries to start with, which I'm particularly fond of.
I've been aiming for -18 LUFS but have recently been happy with my mixes at -20. Have you read the AES paper on recommendations for unifying streaming loudness standards? Its named "Technical Document AES TD1004.1.15-10 “Recommendation for Loudness of Audio Streaming and Network File Playback”
Hi Ken. Yes, good advice from my point of view. The AES recommendations are, in a nutshell: It is recommended that the Target Loudness of the stream not exceed -16 LUFS: to avoid excessive peak limiting, and allow a higher dynamic range in a program stream.1 It is recommended that the Target Loudness of a stream not be lower than -20 LUFS: to improve the audibility of streams on mobile devices. It is recommended that short-form programming (60 seconds or less) be adjusted by constraining the Maximum Short-term Loudness to be no more than 5 LU higher than the Target Loudness: This ensures that commercials and similar short-form content are consistent with the stream loudness. It is recommended that the maximum peak level not exceed −1.0 dB TP: to prevent clipping when using lossy encoders.
Hey Curtis thanks for this. Been a real issue for me. I went back and randomly sampled several of your recent videos and found that most of them were normalized to about -9db (although this one is -2.6). My question is do you have a specific goal for what level you want to achieve for all your videos? Does it change according to subject?
Thanks Dennis. First, note that UA-cam’s numbers are relative to their chosen target of -14 LUFS. So my videos are generally targeted at -17 or -23 LUFS depending on whether I’m reviewing a microphone or recorder (-23 LUFS) or covering something else (-17 LUFS). The reason I target -23 for mic and recorder reviews is I want viewers to hear how they sound without any processing, especially compression, and that is generally possible at -23 LUFS. Once you push the loudness to -16 or -17 LUFS, compression is generally required. Hope that makes sense.
Thank you Curtis for yours recommendation to go with -16 in general. I will define this as Target for our company social media with podcasts and Videos.!
Thank you Curtis!! Everything you say makes a lot of sense to me but now I have a quesiton. Why does spotify crank everything to -14? Because its a different thing when you listen to music or educational podcast shouldnt they have something like a vaiable loudness for different categories? Whats your opinion? Thank you!
Remember that Spotify started as a music streaming service. They did what I think is a wise thing to help manage the loudness wars which have been plaguing the popular music world for a few decades now: Set a reasonable loudness standard of -14 LUFS. That's great for popular music. But in my opinion, it is not as great for material which consists mainly of spoken word content. Spoken word audio tends to sound a little too compressed at -14 LUFS in my opinion. I believe that -16 LUFS is a better middle ground - loud enough for playback on consumer grade equipment in poor listening conditions, but not so loud that there's no dynamic range left. I wish they would use different standards for different categories. I think an argument could be made for orchestral/classical music. That isn't going to sound great at -14 LUFS.
In the home audio world, some of us remember when "loudness" referred to a special EQ control based on the long historied Fletcher-Munson curves that describe human hearing as it relates to the perception of loudness. The usual take-away for mixing engineers was to mix a track so it sounds its best when played back at high volume. Hence the phrase, "crank it up, dude", right? Thanks for the link to the ITU document. Great discussion in it about true peak vs peak-sample metering -- certainly worth being aware of.
Thanks for info, I had aimed at -12 for sometime for internet radio believing tidal/spotify were at that range looks like changed around 2017. Although looks like I am the only one submitting to station with any consistent level as other shows a bit louder or up and down. Nice to try and pretend I'm doing it roughly correct. 😂
Interesting takeaways. I've always wondered if youtube has any standards because when I'm watching on my smart TV I have generally found the loudness of various videos compared to each other is all over the place, forcing me to ride my volume knob every new video.
Audible.com insists that there audiobook files be delivered between -18 and minus 23 RMS seems like a pretty wide allowable range, they only use RMS not the lufs I guess it's something they've been doing a long time and haven't really seen a reason to change although the audio pauses ie blank spaces audio books are typically much shorter or should I say more regular been something like the audio on TV so maybe that makes it more accurate closer to what lufs might appear to be?
Great info! Can't wait for the next one on the topic. Ironically, I watched the video on mute with subtitles on, as my wife is busy working on something on the desk beside mine.
Great and precise explanation of a concept that is often misunderstood. Your last comment in regards to uploading spoken content on UA-cam at a more conservative level is a very wise practice. Thank you.
Well that starts to make sense. In some of your videos you talk about something being set for European LUFS. I thought it was short for Luftballons, but I figured it should have been set for 99. No wonder all of my stuff has been so darned loud.
You can see part II on how to loudness normalize here: ua-cam.com/video/lHNhxGojFRU/v-deo.html
i want to ask.. i record my video on smartphone and i have my record voice in -3 db but the volume on record is on 50-60% and i try to put music to see the bar its on -3 db but the volume its on 80 - 90%... so did i must gain up the volume my mic right? i confuse in here
Nothing's louder than my creaky floors at 3AM
Run them thru an expander.
Perhaps you could use that as your target.
This is the best video ever. Especially the online advice, really been a bit lost as to why -16/-17. Thanks
👍🙏
Excellent video.. Simple explanation, no giff gaff.. Thank you!
👍
As always, this was informative and pleasant to watch. Thanks Curtis!
Thanks Meridian Media!
Great Job as Always! Definitely agree, I always try to normalize my UA-cam uploads to -16 LUFS 👍🏻
👍
Chris, in that upcoming segment, you're gonna need to address amps that go 11 in terms of LUFS. Thanks for the valuable info.
Ah, yes, that'll be in the volume segment. 😉
I expect Curtis to be chewing gum as he tells us they go to eleven, too.
Maybe even wearing a mullet wig...
What a cliff hanger!! ill leave this window open so i dont miss it. thank you for these
👍
Curtis, if I had wanted a mouthful of something, I'd go get a Hardee's #3 double burger...
Good stuff, sir! This is one of the many things that plague newbs on the processing side of things, myself included. Thanks again!
Also, that thumbnail is such a sweet looking piece of work! I particularly love the dichotomy of the colors behind the dial.
Hahaha! Now you made me hungry for a burger!
Kacper, my thumbnail guy, does brilliant work.
I didn't know about this. Just checked my last video and YT has it -9db, so that means I'm around -23? Ready for Netflix!!!
Yes, exactly!
Brilliant explanation and can’t wait for the next instalment. Have learnt so much about great audio from your channel.
Thanks Gareth.
Excellent presentation on an important topic👍
Thanks.
Great information, thanks. Looking forward to more
Thanks Jackson.
This is gold, looking forward for the next video. Thanks a ton for that.
Thanks dlapierre99.
@@curtisjudd I found the Loudness meters in my NLE. They are in LUFS, great. But there are 4 measurements: Momentary, Short, Integrated and Range. I guess that the numbers you gave for YT or Spotify is the integrated. But what about the Momentary and Short, what are good values for those?
@@dlapierre99 Yes, integrated is the measurement of the entire video but you have to play though the entire piece to get the final integrated measurement. The others are for analyzing and solving issues at particular spots within the piece, e.g., maybe I need to compress or level out this little bit here...
@@curtisjudd Thank you so much! I have an option to generate a loudness log during render so I can get the integrated value quite fast if I render only the audio. And I am able to graph each 4 metrics over time in Excel. Thanks a lot again, now I am able to understand what I am doing.
the audio is on another planet
Wait, so if a tree falls on Mars, does it make a sound?
Thanks!
It absolutely does - the rover hears it.
Great explanation, thank you!
Thanks Elliot.
3:39 “It is an old maxim of mine that when you have excluded the impossible, whatever remains, however improbable, must be the truth.”
― Sir Arthur Conan Doyle, The Complete Short Stories of Sherlock Holmes
As a Sherlockian, I got distracted by the quote! Hahaha...
Sherlock!
Excellent, as always.
Thanks Marshall.
Great video; thank you very much!
You are welcome!
This video is great! Thanks Curtis! Typically when I edit a short video that has loads of different sound pieces, I edit each one separately and do a loudness normalize to it. I'm not sure if it is the correct way or I should export the sound as a whole and then make the loudness normalize to it. Will you be touching it on the next video? Thanks again!!
Yes. Here's the short version: Do your mix first. Then loudness normalize the entire mix as the very last step.
@@curtisjudd Thank you! I will do it from now on :)
Hey Curtis, here is a problem I want to solve with audio. My wife and I do UA-cam videos about RV life, We both use lav mics which go into our zoom h1 split into left and right Channels. I do this so I can control our audio separately, I end up having to cut out my Audio when she talks and vice versa. (There is an echo sound of picking up each other’s mic.). This takes a long time. Is there a way to do this automatically in premiere pro? Thanks Larry
Hi Larry, I'm not aware of an automatic way to do this in Premiere (but I haven't used Premiere in several years so perhaps they've added an auto mix feature?)
Much appreciated!
👍
This was a great video!
Thanks!
Assuming this does not affect the recommended recording gain levels of -12db on average?
Correct!
Thanks Curtis, but you mentioned something which doesn't immediately seem related to the topic here - you mention that vocals are *compressed* when normalized to the recommended loudness. It's not clear to me why amplification results in compression? Where does the compression come in?
Hi Richard, that will become clearer in the next video on this topic. Usually one will need to do some dynamics control (compression) to achieve a target which is louder than -23 LUFS for spoken word content. Compression is almost always part of that process or some of the audio transients (short peaks) will be heavily limited.
@@curtisjudd I look forward to the next one! 😁
almost there, hurry up make part 2 lol
👍
Yelp! Going back some time now, on one of your videos you recommended -23 LUFS (LKFS in Resolve) as the level to normalise your audio. I've been doing this for some time now to provide a consistent audio level to all of my videos. Should I now change that to -16?
It depends on where you publish or broadcast. For broadcast TV, they have specific requirements. In the EU, that is -23 LUFS. For the US, -24 LKFS. For online, I generally recommend -16 to -17 LUFS.
@@curtisjudd Thanks. For me it's just online, so it looks like I'm going to have to amend my processes.
Thanks you for mentioning math, but not digging in.. :)
👍
Don't be a victim of the loudness war!
Exactly!
I really needed this information!!! I’ve been looking for this for years!!! Thanks for being so smart!
Thanks for coming by and thanks for teaching us all how to do lighting design!
Same here, been searching for this simplified yet spot on tutorial
@epiclightmedia is that one channel that I had to unsubscribe from due to their excessive amount of helpful high-quality videos that helped me get better ;)
@@gibblsworthiscool All that value is hard to digest... 🙃
I LOVE learning more about hifi audio techniques/voice over, THANK YOU for all this great info Curtis!
👍
Can't wait on the second video on lufs.
👍
Awsome info
Thanks Sufi.
What I do is take live calls on live stream and have six dedicated wattsapp lines..
The audio technic used I have learnt through this your channel which I appreciate and THANXX u loads.. 👍
Fantastic video. I found your channel while looking for a microphone and you've been dedicatedly putting out seriously high quality content for what looks like more than a decade! Props!
Thanks Bjorn.
I'd like to point out that television shows on TV have different technical requirements for audio than the commercials to. That allows TV commercials to be louder than program material.
More info, please? What are the targets for advertisements vs program? Thanks!
@@curtisjudd I'll clarify. Television shows and movies are produced for release on their platforms. Movies may be adjusted before airing to adhere to LUFS/LKFS guidelines too but TV stations often compress any program content. They also visually process shows too which is why TV is a terrible place to watch a movie or TV show. These are done within guidelines (which are minimum specs) to make it a better experience on TV.
Movies and TV shows are produced to look and sound a certain way and if it were possible to air without changing anything, it would be ideal for those of us that worked on it but there are broadcast and cable cast regulations that must be followed. I worked on a show that aired on USA and me and the sound mixer heard all kinds of distortion and compression when the show aired because that's what the station does to all program content. My mixer recorded the show and sent it to post and asked them about some of the things he was hearing. They responded back that their master didn't sound like that and this was common in TV. If we bought the DVD of the show, it would sound better. He was right.
It's been over 2 decades since I worked at a TV station and that was pre-HD but I'm sure stations still have safety measures to make sure audio doesn't go out too loud or that dark scenes don't go too dark.
Commercials, on the other hand, want to be noticed and may do tricks to get attention. Extreme compression is one truck they'll use so that if you're watching a late night show and doze, you'll wake up and watch. Another trick they'll use it playing the A-weighting curve - boosting the crap out of mid and higher frequencies because you hear them better. Really annoying infomercials may even apply aggressive low and high pass filters so that there's more headroom for an annoying presence boost (for example). Commercials don't have to be cinematic masterpieces, their goal is to sell a product so if that means using tricks to sound louder and be noticed, they will. Commercials aren't immune to processing but aren't afraid of playing games.
@@SoundSpeeds Thanks Allen.
@@SoundSpeeds No need to be brief here, sir. Tell me more...
Hahaha, thanks for the wealth of input. Your information coupled with what Curtis conveyed tells me that by sustaining certain loudness levels with very few quiet moments, infomercials and such can be perceived to be louder but still fly below the standards that mainly account for dynamic range differences. Am I understanding you both correctly?
@@insanejughead I think I understand that and I think that's right. Curtis said sustained loudness over time is perceived louder than audio with more dynamics and that's 100% right. What I added is that sometimes commercials and infomercials will use tricks to maintain loudness for the entire time and/or boost frequencies humans have a more difficult time ignoring while cutting those we may tune out.
Thanks for this!
👍
This makes me feel like I'm back in audio school. Loved it. Super clear explanation and really useful. Thanks, Curtis! ~Brian
Thanks Brian! Means a lot coming from you!
Your content is amazing. Thank you so much for this. I've been struggling with audio, particularly when it comes to loudness for a while. Part of that is me, and the fact that I feel the need to news commentator "shout" on camera. But there is also the technical factor of where I should be in terms of audio signals. And this helps with that tremendously. Can't wait for the next video where we learn about tooling.
Thanks Jack!
Always learn something new with You. Thanks
Thanks Leonardo.
This has always been one of the most confusing things about video production. When I made my first TV commercial, they kept rejecting the video for two reasons... the loudness and the luminance / whiteness levels. The same side of two different coins. I wish I saw this before all that trial and error. I would love to have a button in Final Cut Pro that lets me just click a dropdown to use the correct levels for various platforms.
Sadly, I don't think we'll see that in FCPX. It doesn't have the concept of a master audio bus which is where the final loudness normalization would need to take place - unless they make some BIG changes to FCPX.
@@curtisjudd Without industry standards in FXPX the best we can do is use their generic Loudness Amount and Uniformity percentages which I set at 40 and 17 respectively for UA-cam. My newer flow is 32 bit float using Zoom F6 and MixPre 6 II, so I've been doing normalization in Reaper first.
@@AskJoe Yes, makes sense. I do my normalization in Audition or Izotope RX.
Thanks Curtis, happy new year to you! Great video as always! Would you be able to recommend a sweet spot when working with something like very dynamic live orchestral for UA-cam? Thanks
I'd recommend trying different targets to see how they each sound and choose from there. Best if you can get it to -16 LUFS and be satisfied that it doesn't sound too compressed.
@@curtisjudd cheers Curtis!
Curtis, that's amazing, I've just sent a question about LUFS on UA-cam in your online course, and you came with this video:) anyway, the question is - what is the right LUFS level for the UA-cam videos in 2021, for mono and stereo?
Updated: I've reached the end of your video and finally have the answer, thank you! Funny fact from Russia: the "statistics for nerds" was translated by UA-cam as "statistics for system administrators ", so we're lacking some humor there🙃
For popular music, -14 LUFS for stereo, -17 LUFS for mono. My opinion: For spoken word content, I'd aim for -16 LUFS stereo or -19 LUFS mono.
@@curtisjudd And what do you recommend for videos with a mix of spoken word, music and sound design? I think you have to crush the dynamics too much to keep it on -14 LUFS without distortion.
@@Ditispeter -16 LUFS stereo unless the piece will air on broadcast TV.
@@curtisjudd Thanks! I will try that.
@@curtisjudd I always use the --16/-17 LUFS for my UA-cam videos, however I have to practice more in the audio part when I edit and mixing.
Very informative. Thank you.
You're welcome.
You were reading my mind - I was recently wondering about those LUFS and experimenting to get a good level. So, I just changed your 444 likes into 445 ;-)
Thanks Kamil!
Glad to hear the remark at the end, it sounds like you're going to cover the question I was about to ask. I use FCPX with access to LogicProX, and want to know how to estimate my LUFS before I upload the video, ha ha. I always seem to undershoot, compared to professional style videos that somehow get it nearly perfect on stats for nerds.
I won't be able to cover FCPX in particular, but will show my process.
Awesome video, thanks :) Just bought your "Dialogue Sound Post Processing with Adobe Audition" course, do you plan to record a course for Isotope tools? Looks like you're using Resolve instead of Audition for most of the mastering.
Oh, it's already there :) Thanks once more!
Still, I wough HIGHLY recommend adding this information to the course promo/landing page/description. So it's clear for potential buyers that the course is even bigger/better/etc. It would affect sales in a good way for sure.
Thanks Oleksandr. Yes, I do hope to make an Izotope RX course.
Very interesting Curtis. Thanks- And the Stats for Nerds is cool- I had never heard of this before!
👍
You teach me more than my lecturer did. Thank you for sharing this! Can't wait to wait to watch the next episode.
Thanks Ahmad.
Hitting audio standards in Final Cut is frustrating. Or else I need to learn the tools better. In any case, I appreciate it if you would address workflows outside your own Adobe ecosystem.
I totally agree. I finally got frustrated enough with FCP and bought a copy of Izotope RX8 (though there are many other capable tools as well) to normalize all audio before importing audio separate from video. RX8 has output presets for broadcast, streaming, podcast, and more. Adding this step takes a little longer at first, but knowing that my audio levels are correct for output has reduced my level of anxiety with the finished product. No affiliation with the company, just a satisfied user.
Hi Danny, I use FCPX as well. But I don't do the audio processing there for the very reason you cite - it isn't good at that, especially when it comes to hitting a loudness target/standard. I do most of my loudness processing in Izotope because it is fast.
@@curtisjudd I’m eager to watch your technique. I’m an iZotope user also
This is was very interesting. The lighting and sound always on point.
Thanks Paul.
This is great! I’ve always struggled with This. U r awesome
Thanks Paul.
I've only been following you for about a year, Curtis, but I've learned so much in that year. Thank you not only for providing your insights and knowledge but also for giving those insights "handles" and showing us how to use them.
👍
Great comment.
So helpful, thank you! Finally know what lufs is haha
👍
So my question would be in Streamlabs what should be my dB target at any given time that would most closely correlate to - 14 LUFS, - 16 LUFS and lastly - 17 LUFS over time? I get that they aren't quite the same thing but I believe there should be a target dB level that if maintained most of the time that should correlate to a given LUFS, right?
During my live streams I have game audio from a game like Call of Duty Cold War, music from Epidemic at 320 kbps and my voice from me speaking into a SM7B all mixed together in real time via my GoXLR mixer. So there usually is a pretty constant stream of audio sounds as one of the above is almost always playing, usually several or all even at once.
Unfortunately, as we demonstrated, dB RMS doesn't always work out. That's the closest alternative of which I am aware.
Is there a way to do Loudness Normalization in Premiere itself instead of taking the route through Wavelab/Audition?
I end up using so many different softwares for my videos so I'm trying not to add more steps ;)
Yes, when you export, go to the effects area and choose loudness normalization.
I have been doing youtube stuff for nearly eight years, I am better with my audio than most, because it's the first thing to bug me (and many, many others) in a video. But this was very helpfull, thank you!
Glad it was helpful!
Just learned there is a right click to get 'Stats for Nerds'!.
But only click on that if you actually are an audio nerd! 😀
@@curtisjudd Well, I have only you to blame for that. Last few months, you have made me deep dive into audio and get familiar with tools and technology I did not know existed.
@@TechforToastmasters Welcome to the nerd club, nerd. 😎😁
Now you're stuck in this rabbit hole with myself and the rest of us. We're a friendly lot, though...
@@insanejughead Indeed. When I first started, I thought buying the best microphone would solve all my problems so I went and bought a SM7b. That could not be further from the truth. Now my room is fitted with thick carpet, lots of soft furniture to dampen the sound, and paying close attention to levels, clipping, compression, equalizer, etc.
@@TechforToastmasters I don't have the SM7b, but I do have two Unidyne III 545's. The information I have found implies that the capsules are the same between the classic 545, SM57, and SM7b, except that the housing and transformer is different to create a subtly different sound for each.
Actually, isn't the SM7b transformer-less? Otherwise it wouldn't need so much external preamp gain, right?
I always learn with your vids tx
Thanks!
I just checked my most recent upload, and my loudness is only -0.4db, not bad!
Very good, in fact!
Curtis, Thank You again for sharing so much knowledge, we really appreciate it! Happy New Year to you and your family from Spain!
Thanks and same to you and yours!
This is why I subscribed many years ago mr Judd. An inspiring video. Cheers!
Thanks Edo.
Curtis, you are the MAN. Do you have any idea how much I've learned from you, and how much those lessons have had real life changes in my workflow and understanding what is going on, so I can make better choices? Thank you a million times over.
Thanks so much, David!
The recommendation is -16 LUFS for YouYube with primarily voice. Got it. What happened to -23or -24 LUFS? Only for broadcast? BTW, saw the normalizing video. fantastic. Thanks!
-23 LUFS is the EBU target for broadcast television. -24 LKFS is the ITU target and is required for broadcast TV in the US. The internet is the Wild West with no requirements. My recommendation is -16 LUFS for online content.
Thank you so much for all your content and excellent delivery and production quality. Straight to the point (no BS) and textual summaries to start with, which I'm particularly fond of.
Thanks Henrik. 👍
I've been aiming for -18 LUFS but have recently been happy with my mixes at -20. Have you read the AES paper on recommendations for unifying streaming loudness standards? Its named "Technical Document AES TD1004.1.15-10 “Recommendation for Loudness of Audio Streaming and Network File Playback”
Hi Ken. Yes, good advice from my point of view. The AES recommendations are, in a nutshell:
It is recommended that the Target Loudness of the stream not exceed -16 LUFS: to avoid excessive peak limiting, and allow a higher dynamic range in a program stream.1
It is recommended that the Target Loudness of a stream not be lower than -20 LUFS: to improve the audibility of streams on mobile devices.
It is recommended that short-form programming (60 seconds or less) be adjusted by constraining the Maximum Short-term Loudness to be no more than 5 LU higher than the Target Loudness: This ensures that commercials and similar short-form content are consistent with the stream loudness.
It is recommended that the maximum peak level not exceed −1.0 dB TP: to prevent clipping when using lossy encoders.
This is awesome, thank you
Thanks.
Great video. Happy New Year. Looking forward for the next video you mentioned how we practically achieve LUFS for UA-cam!
👍
Thanks for this video! Looking forward to part 2. I’ve been unsure of what audio levels to shoot for in premiere for years
👍
Thank you
👍
Hey Curtis thanks for this. Been a real issue for me. I went back and randomly sampled several of your recent videos and found that most of them were normalized to about -9db (although this one is -2.6). My question is do you have a specific goal for what level you want to achieve for all your videos? Does it change according to subject?
Thanks Dennis. First, note that UA-cam’s numbers are relative to their chosen target of -14 LUFS. So my videos are generally targeted at -17 or -23 LUFS depending on whether I’m reviewing a microphone or recorder (-23 LUFS) or covering something else (-17 LUFS). The reason I target -23 for mic and recorder reviews is I want viewers to hear how they sound without any processing, especially compression, and that is generally possible at -23 LUFS. Once you push the loudness to -16 or -17 LUFS, compression is generally required. Hope that makes sense.
Thank you Curtis for yours recommendation to go with -16 in general. I will define this as Target for our company social media with podcasts and Videos.!
👍
Thank you Curtis!! Everything you say makes a lot of sense to me but now I have a quesiton. Why does spotify crank everything to -14? Because its a different thing when you listen to music or educational podcast shouldnt they have something like a vaiable loudness for different categories? Whats your opinion? Thank you!
Remember that Spotify started as a music streaming service. They did what I think is a wise thing to help manage the loudness wars which have been plaguing the popular music world for a few decades now: Set a reasonable loudness standard of -14 LUFS. That's great for popular music. But in my opinion, it is not as great for material which consists mainly of spoken word content. Spoken word audio tends to sound a little too compressed at -14 LUFS in my opinion. I believe that -16 LUFS is a better middle ground - loud enough for playback on consumer grade equipment in poor listening conditions, but not so loud that there's no dynamic range left. I wish they would use different standards for different categories. I think an argument could be made for orchestral/classical music. That isn't going to sound great at -14 LUFS.
I hope you show how YOU achieve Television Loudness EU -23 LUFS, USA-24 LKFS with the tools you own.
That is exactly what I'll do.
In the home audio world, some of us remember when "loudness" referred to a special EQ control based on the long historied Fletcher-Munson curves that describe human hearing as it relates to the perception of loudness. The usual take-away for mixing engineers was to mix a track so it sounds its best when played back at high volume. Hence the phrase, "crank it up, dude", right?
Thanks for the link to the ITU document. Great discussion in it about true peak vs peak-sample metering -- certainly worth being aware of.
Thanks
He hits me with a tongue twister @4:01 then leaves me with a cliff hanger! Thank goodness I'm subscribed! Awesome Lesson! Well done!
-Will
Hahaha! Thanks Will!
neat
Thanks.
Thanks for info, I had aimed at -12 for sometime for internet radio believing tidal/spotify were at that range looks like changed around 2017. Although looks like I am the only one submitting to station with any consistent level as other shows a bit louder or up and down. Nice to try and pretend I'm doing it roughly correct. 😂
😀
Interesting takeaways. I've always wondered if youtube has any standards because when I'm watching on my smart TV I have generally found the loudness of various videos compared to each other is all over the place, forcing me to ride my volume knob every new video.
Yes, turns out that they only enforce the upper loudness limit (-14 LUFS), and not anything that falls below that.
That's clear, Curtis. Thanks for sharing. Happy New Year. :)
Thanks Rog Berube! Happy new year to you!
I had no idea that the stats for nerds existed at all. Don't forget to smash the like button on this video.
Thanks Musical Kinetics.
Audible.com insists that there audiobook files be delivered between -18 and minus 23 RMS seems like a pretty wide allowable range, they only use RMS not the lufs I guess it's something they've been doing a long time and haven't really seen a reason to change although the audio pauses ie blank spaces audio books are typically much shorter or should I say more regular been something like the audio on TV so maybe that makes it more accurate closer to what lufs might appear to be?
I think they just haven't updated their specs. -18 to -23 dB RMS is a potentially huge range.
You’re the man! Best way to learn audio from the master. Thank you, Curtis 😊
🙏
Great info! Can't wait for the next one on the topic. Ironically, I watched the video on mute with subtitles on, as my wife is busy working on something on the desk beside mine.
Hahaha! Good not to disturb the wife!
This was SUPER helpful. Thanks to you and your team!
Thanks Paul!
Extremely helpful, Thank you so much!
👍
Embarrassed to say how long this has been in my Watch later playlist.... Glad I finally watched it tho. :)
Hope it was helpful!
Great and precise explanation of a concept that is often misunderstood. Your last comment in regards to uploading spoken content on UA-cam at a more conservative level is a very wise practice. Thank you.
Thanks Daniel.
well explained, thank you sir
Thanks.
Will definitely need to work on this on my future videos! Thanks Curtis!
You're welcome and best wishes!
Thank so mutch, I love this Video. I think every pro video creater should use this method.
👍
Awesome useful info thanks. Was wondering difference btw the two
👍
More people should switch to Tidal, their compression is the most minimal, they have a ton of high def music.
Thanks for that. Do you know what they target in terms of loudness?
Stats for nerds on this vid: -2.6dB. Is it too late to write-in Curtis Judd for president?
Hahaha! -16.6. Pretty good!
Why couldn't you be my teacher at school?
Hahaha! Thanks.
awesome vids
Thanks.
Well that starts to make sense. In some of your videos you talk about something being set for European LUFS. I thought it was short for Luftballons, but I figured it should have been set for 99. No wonder all of my stuff has been so darned loud.
Hahaha! Definitely do NOT aim for a target of 99 LUFS.
V nice. Tbx. Looking forward to the follow up.
Thanks Phil.
How can we measure the loudness and the volume of a musical instrument.. thanks
Same as covered here - need a loudness meter. A free option is the Youlean loudness meter.
This is a very excellent video. Thank you for posting it.
You’re most welcome, Dan.
Thank you! This will help so much with the levels that I have to deal with with customer's videos.
👍