Thanks Curtis !! I was looking for a limiter in RX and couldn't unbelievably find one. You learned me they were hiding in the plug-in module !!! You spare me a thousands curses more than I already said :D
@@curtisjudd I wonder if you know how to best replicate your final loudness controls in RX7 Essentials? There is no way to just set limit and desired volume and render. I currently use mix of Gain and Normalise and analysing the track after each pass. Is there a better way? Thank you.
@@moristar Hi Alex, depending on the DAW or NLE you're using, some of them have ways to automate loudness normalization. Audition and Premiere, for example, have the ability to match loudness standards - match loudness panel in Audition, loudness normalize during export in Premiere.
Fantastic tutorial - confirmed I am correct in removing asymmetry first (via Phase module), introduced me to amplitude zones and (how to use) frequency bands (for this given context) in compression. Thanks enormously.
Hey mate, just wanted to say thanks for the informative video! I've been in pro audio for 5+ years and never knew about the asymmetric peaks/troughs and the benefits of correcting them with the phase module. Love learning about this kind of stuff =)
Great video. I watched your course on Audition, and it cleared up a lot of question on the workflow. I hope you do a similar course on the entire editing flow in RX 5
Thank you very much for this. Finally someone explaining audio editing in a mathematical way I can understand. I was very disappointed when we got to the compression part, I have the same plug-ins as you do (RX6 and Ozone 7 on my end) and compression has always been my biggest hang up, so when you said "we're gonna just go through this quickly" I was like "nooooo!" I'm still fiddling with the controls trying to get the Integrated part high enough to lower it to 19, but nothing is working, even matching up with the peaks like you did only gave me about a .2 increase. But thank you for the content you did give, it makes sense to me now how to process to -19 up til the compression part which is more than I had before. I'll do some more homework and see if I can figure it out.
Hi Curtis, I tried doing what you did, and then moved the threshold down, and keeping the limiter part at -6 since that made sense to me to not have it go over that (still trying to understand it all), but then I was in a weird loop where anything I did with the Threshold seemed to make the True Peak a larger negative number, when I was just trying to fine tune it, so i had to start over. I'll watch your video on In-depth Compression to see if I can understand what I'm doing more, and maybe figure out the limiter side of things. Thank you for the response, I appreciate it!
This is great video. I learned a couple of tips that I've started using in my audiobook recordings that have really helped. One problem though. I downloaded the VST compressor plugin for use in RX5 (not advanced). I've pointed RX5 to the plugin and it won't open. Somewhere I read where Izotope has "black-flagged" some VST plugins. How do I get RX5 to recognize and open the compressor plugin? Should I use the 32- or 64-bit plugin?
Hi Jughead Jones, I would go with the 64 bit version of the plugin. I'd also contact Izotope support - they should be able to help troubleshoot, especially if Izotope is blocking the plugin for some reason. Best wishes!
Hi Curtis and thanks very much for this video. It helped me to understand a few things I wondered about before. 3 Questions: 1. In this video you are using a mono audio file but can all this apply to a stereo file as well? 2. And, if you apply the Adaptive Phase Rotation to a stereo file, might it upset phase relationships within the file itself and the phase relationship between itself and other audio files in a DAW? 3. As long as it's used lightly, would a limiter be ok to use verses a compressor? I ask this because I use a normalizing plug-in that uses a limiter instead of a compressor. RX 7 at $1000 is a bit too expensive for me at this point. Thanks, Steve
Hi Steve, 1) Yes 2) It hasn't made anything worse in my experience 3) Yes, a limiter can work as long as it has a bit of a soft knee and you don't need to manage super extreme spikes/transients. Best of luck!
@@LearnLightAndSoundSessions Thank you Curtis and take care. By the way, I actually studied the Zone System in college way back when. It was a good foundation for other disciplines in the media business.
Great discussion, I learned a lot. Unfortunately Rx5 Loudness tool is only in the Advance version. The standard version does not have this capability. Are there any plug ins that would have the same loudness capability?
Thanks Gregg. Good catch, I thought the non-advanced version had the loudness module but you are correct. You can also do this in Audition which includes both real-time loudness measurement plugins (Loudness Radar by tcelectronic - top notch) and an offline amplitude statistics panel which also measures loudness and true peak amplitude. The main thing that Audition does not have which I demonstrated here is an automated phase rotation plugin. It does have a stereo phase tool which takes a bit of work and the results aren't usually quite as nice.
I know mainly because my voice is highly asymmetric. I thought something was broken for the longest time. turns out that its just a "feature" of some peoples' voices.
Your videos are outstanding, Curtis. Got a workflow question for you. Would you process audio before or after nornalization? For example, if I were to tweak with some EQ, whould you do it before or after? Also, you compressed it a couple of times. What advantage does that have over a hard limiter? Thanks for sharing so much info.
Hi and thanks! Some before and some after. My main goal with compression of dialogue is to pull the transient peaks down so that I have enough headroom to normalize. My goal is usually to NOT change the sound of the dialogue. I use compression instead of just a brick wall limiter because in many cases the transients can start to sound odd when chopped off with a limiter. With EQ, depending on what you're trying to accomplish (fix issues vs. sweeten) it may make sense to do issue fixing before and sweetening after. We cover much of this in a bit more detail in my post-processing course at school.learnlightandsound.com
Great informative video, thank you! Can I ask what the target for a stereo spoken voice would be? Is it the same as you mentioned for mono voice: -1.5db peak and -19 LUFS.
I installed the Melda Production free plugin bundle, but i can't seem to find it in RX6! It can see all my other plugins from those folders. Please help!
Run the installer again, and pay attention to install folders, there is differents folders for 32 or 64 bits, you have to add the folder in rx7 plugin preferences, my folder of 64 bits plugin is "C:\Program Files\Steinberg\VSTPlugins\MeldaProduction"
Great video!!!! thanks for sharing your knowledge!!. Curtis, Shall I normalize track by track ? Do I have to normalize audio at the end of the procees when all my audio is edited?. I cleaned an audio track but normalizing brought the noise floor up :(.
Hi, yes, normalizing will ALWAYS bring the noise floor up. In a mix, you will typically cut out any "silent" sections from the dialogue tracks and just leave room tone (or the room tone/noise floor from one dialogue track and cut all the others). I usually just loudness normalize at the end once the edit and mix are done.
Thanks for this, Curtis. I'm lucky in that I know some professional dubbing mixers that sort out my post-audio but I wouldn't mind starting to learn a bit to make myself more self-sufficient. Moving up from something like Audacity, would you recommend this or Adobe Audition?
Audition is definitely more flexible and general purpose whereas Izotope RX is a specialist tool which is really great at cleaning tasks. I'd probably start with Audition. Thanks!
Asymmetry is not unusual: "many acoustic sources inherently have a 'positive air pressure bias' because of the way the sound is generated. To talk or sing, we have to breathe out, and to play a trumpet, we have to blow air through the tubing. So, in these examples, there is inherently more energy available for the compression side of the sound wave than there is for the rarefaction side" - www.soundonsound.com/sound-advice/q-why-do-waveforms-sometimes-look-lop-sided
Thanks for the video. At -19 LKFS, my dialogue clip sounds pretty quiet, and a true peak of -4dB is the best I can do, since my recording started off at -15 LKFS and 0dB true peak. Should I increase the LKFS until the true peak is at -1.5dB, or is the LKFS the most important parameter?
Hi Ksawery, the LKFS/LUFS is the most important target. If you're trying to compete with online streaming music, then by all means, go to -15 LKFS or even -14, which is what I believe spotify and some of the other music streaming services target. For dialogue, I find that many voices sound too crushed at -14 or -15 LKFS so I go with the more conservative -16 LKFS recommendation.
Thanks for your quick reply. I've been doing some more research today, and the consensus seems to be -16 LKFS for UA-cam, as you say, although apparently you are allowed to push it to -13 LKFS (but I think that would require too much compression). I think -16 LKFS sounds good and I'll stick to it from now on. Thanks!
I wonder if the mentioned dialogue clip was stereo. As Curtis explains elsewhere in these comments, -19 LUFS is for a mono clip, -16 for a stereo clip. (Incidentally, he reasoning for the difference being 3 dB is not quite as simple as one might imagine, but it is correct).
Hey Curtis, would you say most video people are doing there audio processing destructively (like you show in this vid) or in real time with plugins in there NLE's? (Audio DAW's use processing slots in the audio tracks that you insert real time processors that are categorized as "plug-ins") I havnt used NLE's other than Davinci Resolve. (which doesn't have the ability to use industry standard plugins like the Ozone stuff.)
Hmm, not sure. Offline processing, which is usually "destructive" is pretty much necessary for things like loudness normalizing I think. But I'm open to learn better ways if you've got some input! :)
This tutorial seems a little too advanced for me. I'm in need of one that carefully runs me through WHY each step is there and exactly what it is doing in FULL CONTEXT of the process. Totally my fault, not this video. Anyone know a good video for that?
Try these two. Loudness vs Volume: ua-cam.com/video/nOzZKEWJ5wk/v-deo.html How to Loudness Normalize Your Audio: ua-cam.com/video/lHNhxGojFRU/v-deo.html
Agreed. Any recommendations for managing asymmetric waveforms on mono dialogue recordings. Phase invert simply moves the asymmetry to the other side of minus infinity line.
Mono recordings are easily corrected with Waves InPhase plug-in (can be purchased solo at Waves for small money) or MAutoAlign accordingly - this is probably shown most for multi microphone recordings. I've tested different options for microphone amplifiers, and now using Focusrite Voicemaster Pro - great value for money, and phase correct, and then correction is not necessary :)
Thanks for the feedback Ziyed. This is my alternate channel that is associated with my online courses. These are longer, unscripted sessions so yes, I talk a lot here. If you want shorter, more focused sessions that go less into depth, please see my other channel which is titled, "Curtis Judd' rather than "Curtis Judd Audio".
If I could give this video 100 thumbs up I would! What a fantastic tutorial - now I can get my dialogue to the correct standard!
Thanks, glad you found it helpful!
Thanks Curtis !! I was looking for a limiter in RX and couldn't unbelievably find one. You learned me they were hiding in the plug-in module !!! You spare me a thousands curses more than I already said :D
Haha! Glad it helped!
Best routine for the dialog fixes. Thank you.
👍
@@curtisjudd I wonder if you know how to best replicate your final loudness controls in RX7 Essentials? There is no way to just set limit and desired volume and render. I currently use mix of Gain and Normalise and analysing the track after each pass. Is there a better way? Thank you.
@@moristar Hi Alex, depending on the DAW or NLE you're using, some of them have ways to automate loudness normalization. Audition and Premiere, for example, have the ability to match loudness standards - match loudness panel in Audition, loudness normalize during export in Premiere.
Fantastic tutorial - confirmed I am correct in removing asymmetry first (via Phase module), introduced me to amplitude zones and (how to use) frequency bands (for this given context) in compression. Thanks enormously.
You bet, thanks for the feedback!
Hey mate, just wanted to say thanks for the informative video! I've been in pro audio for 5+ years and never knew about the asymmetric peaks/troughs and the benefits of correcting them with the phase module. Love learning about this kind of stuff =)
Thanks McNougat!
Always great to wake up to a new post from you. Very helpful! Thank you for all the time it must take to do these posts. Very much appreciated!
Thanks Sean!
Thank you. So helpful!
Great video, Curtis! Learned a lot from it, thanks.
Thanks!
Have a question: you mention in the video that -19db is somehow ideal for a mono track. What would be a good level for a stereo track?
-16 LUFS for stereo.
Great video. I watched your course on Audition, and it cleared up a lot of question on the workflow. I hope you do a similar course on the entire editing flow in RX 5
Thanks Mark. A brief course on RX may be in the future. :)
Thank you very much for this. Finally someone explaining audio editing in a mathematical way I can understand. I was very disappointed when we got to the compression part, I have the same plug-ins as you do (RX6 and Ozone 7 on my end) and compression has always been my biggest hang up, so when you said "we're gonna just go through this quickly" I was like "nooooo!"
I'm still fiddling with the controls trying to get the Integrated part high enough to lower it to 19, but nothing is working, even matching up with the peaks like you did only gave me about a .2 increase. But thank you for the content you did give, it makes sense to me now how to process to -19 up til the compression part which is more than I had before. I'll do some more homework and see if I can figure it out.
Hi Phillip, it sounds like maybe you need to adjust the threshold down so that the compressor squashes the audio peaks more.
Hi Curtis, I tried doing what you did, and then moved the threshold down, and keeping the limiter part at -6 since that made sense to me to not have it go over that (still trying to understand it all), but then I was in a weird loop where anything I did with the Threshold seemed to make the True Peak a larger negative number, when I was just trying to fine tune it, so i had to start over. I'll watch your video on In-depth Compression to see if I can understand what I'm doing more, and maybe figure out the limiter side of things. Thank you for the response, I appreciate it!
6:18 Discussion begins about loudness control. (Just a note for my own reference so I can find it quickly!)
8:26 How to use Loudness Control module
9:31 Determining how much compression you need
👍
Great RX tutorials are a +++++
Thanks Felipe!
Gee you make good videos Curtis-- thanks again, SUPER helpful.
Thanks James.
So helpful! Thanks Curtis!
Thanks!
This is great video. I learned a couple of tips that I've started using in my audiobook recordings that have really helped. One problem though. I downloaded the VST compressor plugin for use in RX5 (not advanced). I've pointed RX5 to the plugin and it won't open. Somewhere I read where Izotope has "black-flagged" some VST plugins. How do I get RX5 to recognize and open the compressor plugin? Should I use the 32- or 64-bit plugin?
Hi Jughead Jones, I would go with the 64 bit version of the plugin. I'd also contact Izotope support - they should be able to help troubleshoot, especially if Izotope is blocking the plugin for some reason. Best wishes!
Hi Curtis and thanks very much for this video. It helped me to understand a few things I wondered about before.
3 Questions:
1. In this video you are using a mono audio file but can all this apply to a stereo file as well?
2. And, if you apply the Adaptive Phase Rotation to a stereo file, might it upset phase relationships within the file itself and the phase relationship between itself and other audio files in a DAW?
3. As long as it's used lightly, would a limiter be ok to use verses a compressor? I ask this because I use a normalizing plug-in that uses a limiter instead of a compressor. RX 7 at $1000 is a bit too expensive for me at this point.
Thanks,
Steve
Hi Steve, 1) Yes 2) It hasn't made anything worse in my experience 3) Yes, a limiter can work as long as it has a bit of a soft knee and you don't need to manage super extreme spikes/transients. Best of luck!
@@LearnLightAndSoundSessions Thank you Curtis and take care. By the way, I actually studied the Zone System in college way back when. It was a good foundation for other disciplines in the media business.
@@newguy6935 I agree! The zone system is still very applicable from my point of view.
great video!👍
Thanks!
Great discussion, I learned a lot. Unfortunately Rx5 Loudness tool is only in the Advance version. The standard version does not have this capability. Are there any plug ins that would have the same loudness capability?
Thanks Gregg. Good catch, I thought the non-advanced version had the loudness module but you are correct. You can also do this in Audition which includes both real-time loudness measurement plugins (Loudness Radar by tcelectronic - top notch) and an offline amplitude statistics panel which also measures loudness and true peak amplitude. The main thing that Audition does not have which I demonstrated here is an automated phase rotation plugin. It does have a stereo phase tool which takes a bit of work and the results aren't usually quite as nice.
It's great that you talked about the phase rotation correction. I had no idea some voices are asymmetric in nature.
I know mainly because my voice is highly asymmetric. I thought something was broken for the longest time. turns out that its just a "feature" of some peoples' voices.
Your videos are outstanding, Curtis. Got a workflow question for you. Would you process audio before or after nornalization? For example, if I were to tweak with some EQ, whould you do it before or after? Also, you compressed it a couple of times. What advantage does that have over a hard limiter?
Thanks for sharing so much info.
Hi and thanks! Some before and some after. My main goal with compression of dialogue is to pull the transient peaks down so that I have enough headroom to normalize. My goal is usually to NOT change the sound of the dialogue. I use compression instead of just a brick wall limiter because in many cases the transients can start to sound odd when chopped off with a limiter. With EQ, depending on what you're trying to accomplish (fix issues vs. sweeten) it may make sense to do issue fixing before and sweetening after. We cover much of this in a bit more detail in my post-processing course at school.learnlightandsound.com
Great informative video, thank you!
Can I ask what the target for a stereo spoken voice would be? Is it the same as you mentioned for mono voice: -1.5db peak and -19 LUFS.
Thanks Will. For stereo, the true peak should be -1 to -1.5dBTP and -16 LUFS for online/mobile video.
Great! Thanks Curtis.
What's the difference between loudness normalize and Loundness control?
"Loudness Control" is the name of the module in Izotope RX which one uses to loudness normalize an audio clip or mix.
I installed the Melda Production free plugin bundle, but i can't seem to find it in RX6! It can see all my other plugins from those folders. Please help!
Probably best to contact Izotope support. Best wishes!
Run the installer again, and pay attention to install folders, there is differents folders for 32 or 64 bits, you have to add the folder in rx7 plugin preferences, my folder of 64 bits plugin is "C:\Program Files\Steinberg\VSTPlugins\MeldaProduction"
Great video!!!! thanks for sharing your knowledge!!. Curtis, Shall I normalize track by track ? Do I have to normalize audio at the end of the procees when all my audio is edited?. I cleaned an audio track but normalizing brought the noise floor up :(.
Hi, yes, normalizing will ALWAYS bring the noise floor up. In a mix, you will typically cut out any "silent" sections from the dialogue tracks and just leave room tone (or the room tone/noise floor from one dialogue track and cut all the others). I usually just loudness normalize at the end once the edit and mix are done.
Curtis Judd thanks Curtis!! I do appreciate you spend time to reply these kind of messages. 🙏🏻.
God bless you.
Thanks for this, Curtis. I'm lucky in that I know some professional dubbing mixers that sort out my post-audio but I wouldn't mind starting to learn a bit to make myself more self-sufficient. Moving up from something like Audacity, would you recommend this or Adobe Audition?
Audition is definitely more flexible and general purpose whereas Izotope RX is a specialist tool which is really great at cleaning tasks. I'd probably start with Audition. Thanks!
Asymmetry is not unusual: "many acoustic sources inherently have a 'positive air pressure bias' because of the way the sound is generated. To talk or sing, we have to breathe out, and to play a trumpet, we have to blow air through the tubing. So, in these examples, there is inherently more energy available for the compression side of the sound wave than there is for the rarefaction side" - www.soundonsound.com/sound-advice/q-why-do-waveforms-sometimes-look-lop-sided
Thanks for the resource and background info!
Thanks for the video. At -19 LKFS, my dialogue clip sounds pretty quiet, and a true peak of -4dB is the best I can do, since my recording started off at -15 LKFS and 0dB true peak. Should I increase the LKFS until the true peak is at -1.5dB, or is the LKFS the most important parameter?
Hi Ksawery, the LKFS/LUFS is the most important target. If you're trying to compete with online streaming music, then by all means, go to -15 LKFS or even -14, which is what I believe spotify and some of the other music streaming services target. For dialogue, I find that many voices sound too crushed at -14 or -15 LKFS so I go with the more conservative -16 LKFS recommendation.
Thanks for your quick reply. I've been doing some more research today, and the consensus seems to be -16 LKFS for UA-cam, as you say, although apparently you are allowed to push it to -13 LKFS (but I think that would require too much compression). I think -16 LKFS sounds good and I'll stick to it from now on. Thanks!
I wonder if the mentioned dialogue clip was stereo. As Curtis explains elsewhere in these comments, -19 LUFS is for a mono clip, -16 for a stereo clip. (Incidentally, he reasoning for the difference being 3 dB is not quite as simple as one might imagine, but it is correct).
I don't see Loudness option in RX Element 7
Unfortunately I don't believe it is included in the Elements version.
Why use Loudness (module) at the beginning rather than Normalize ? Latter would surely avoid need for mental arithmetic.
Fair point. Good to keep the mind sharp. 😀
Hey Curtis, would you say most video people are doing there audio processing destructively (like you show in this vid) or in real time with plugins in there NLE's? (Audio DAW's use processing slots in the audio tracks that you insert real time processors that are categorized as "plug-ins") I havnt used NLE's other than Davinci Resolve. (which doesn't have the ability to use industry standard plugins like the Ozone stuff.)
Hmm, not sure. Offline processing, which is usually "destructive" is pretty much necessary for things like loudness normalizing I think. But I'm open to learn better ways if you've got some input! :)
Curtis Judd what r u using for a nle?
Mostly Final Cut Pro X but some work in Premiere and even a little in Resolve. However, I do my audio work in Izotope RX Audio Editor and Audition.
This tutorial seems a little too advanced for me. I'm in need of one that carefully runs me through WHY each step is there and exactly what it is doing in FULL CONTEXT of the process. Totally my fault, not this video. Anyone know a good video for that?
Try these two. Loudness vs Volume: ua-cam.com/video/nOzZKEWJ5wk/v-deo.html How to Loudness Normalize Your Audio: ua-cam.com/video/lHNhxGojFRU/v-deo.html
@@curtisjudd You flopping legend. Cheers!
Its better to get rid of phase problems before you record!
Agreed. Any recommendations for managing asymmetric waveforms on mono dialogue recordings. Phase invert simply moves the asymmetry to the other side of minus infinity line.
Mono recordings are easily corrected with Waves InPhase plug-in (can be purchased solo at Waves for small money) or MAutoAlign accordingly - this is probably shown most for multi microphone recordings. I've tested different options for microphone amplifiers, and now using Focusrite Voicemaster Pro - great value for money, and phase correct, and then correction is not necessary :)
man you talk too much
Thanks for the feedback Ziyed. This is my alternate channel that is associated with my online courses. These are longer, unscripted sessions so yes, I talk a lot here. If you want shorter, more focused sessions that go less into depth, please see my other channel which is titled, "Curtis Judd' rather than "Curtis Judd Audio".