Thanks for this tutorial, we've just started dabling in RX10 Advanced from RX7 Elements, we found this tutorial really helpful and its opened up a whole new learning curve, but very helpful thank you
I've embarassed myself today trying to get a gig as a VA Mastering dude for a Mod (No money involved), they couldn't take me because I didn't know what I was doing with Dialogue, since I've been learning how to work with music, not dialogue for films or videogames, this is a good insight into what is possible with RX, and I feel like with all that content out there, there are just a few that really explain how this software works, like this video. I'll do my best at getting the gig for my own good, I need that confidence for my future.
To clean up that type of filtered broadband noise I use Spectral DeNoise in a specific way (that's actually the only way I use that module). After you hit learn you'll notice that the range of that noise sticks out in the Noise Profile curve. Make sure Reduction Curve is enabled, grab that curve (the blue line) and drag it all the way up for the entire frequency range. Now add anchor points on the blue line just outside that filtered noise that's poking out and drag down enough points in between so that the Residual Noise curve (in yellow) looks smoother. Set your Reduction slider to taste and hit process. Give it a try, that usually does the trick. 😉
I had always learned that numerous processes being done to a single recording degrades it. It sounds like you have gotten better results doing several small processes...
Great video! Thank you, Thomas! Regarding broadband noise, I've found some interesting results with a very (-2/-3db) DX isolate pass on everything and then layer DX isolate for above 700-1khz and then Deconstruct below 300-700hz. Seems isolate retains the airiness of the dialogue whereas deconstruct will think it's noise. Deconstruct down below seems to find that noise between the harmonics better whereas isolate will not. For Deconstruct I'll do 0db tonal, -3/-6 noise, and opt to keep transient separate. Then leave sliders midway. Can work well for noise that often isolate and spectral denoise leave behind.
this tool is great...I am new to audio editing...one of the challenges I have is whether or not to keep or remove noise...the dialogue becomes clearer however it sounds less natural...I was recording a speaker outside during a protest...letting the noise-people bleed into the dialogue gives it character even though the odd word here and there is harder to hear....
The lower frequency hum I'd surgically remove because it's so prominent. Highlight the quiet area a la 1:41, pull up the spectrogram and move your mouse over each frequency (look for the obvious humps), go back to De-hum, choose static like you had in 2:02, but drop the the harmonics to 1 (not 16) to give you the ability to surgically removed each hum frequency. For lower frequencies, widen Q, e.g., 200-500. Rinse/repeat for the other lower frequencies. For the remainder, e.g., above 500Hz, just use spectral denoise, settings C or D.
Great instructional! But do you have a video with tips for getting the best “down & dirty “ results? My problem is dealing with tight deadlines & panicky clients so I seldom have the time to do such a beautifully detailed job like your example.
This small steps approach yields impressive resultats indeed. As you mentioned, I'm one of those users going way harder on knobs, and not always happy witht the results. Intuitively I would have thought that with your approach, artefacts created at each small steps would add up and interfere with each other, even more distorted going through the following step, so the end result would not be good, at least not better than a 1 time carefully set pass. Seems it isn't the case. Thanks for the tutorial. One question : Could you polish the end result, which sounds a bit mids heavy maybe (be it in volume or in harmonic content) to me, by either EQing a bit, or use the spectral repair ? Thank you for this further info
Hi, Do you have a video on how you would remove a cell phone ringtone ( Or anything similar ) from something like interview audio . I have RX 10 Standard. Not sure how much that would call for a different workflow in this case. Thanks
It was the default Motorola "Hello Moto" musical type ringtone. So the ringtone intro is different from the body . It's not just a recurrent tone. . I hope I explained that properly. Not sure what the general term would be.
Thanks for taking your time to do those videos. they are really helpful for new and experienced people in post production sound. You mention resonance frequencies, do you find it helpful to use the eq to remove those after Noise reduction? I ask because removing resonance with EQ can be time consuming. What are your thoughts?
Nice video. The frenetic mouse movements make it a little hard to watch but I learned some good tips. Why did you decide against spectral de-noise? Using Learn on that seems like it would have done a good job with the background noise.
i love your voice lol, dude i hate it how i like it lollllllllllllllllll. Imagine just a random stranger on earth, you being all serious, straight to point, and me just nodding and being like ah such a pleasure to listen to lollllllllll (no homo even if im a girl, just wanted to be honest, dont mean anything by it haha)
Have you tried Waves Clarity VX Pro yet? On some things it reduces a whole RX session into 1 knob. On others it annihilates breaths and anything not distinctly dialogue. Been using it since Thanksgiving on a feature and its potential is huge. Had a lot of noisy birds in a scene and it did miles better than Absentia DX's bird reducer. I'm curious to see how iZotope responds to Clarity.
hello, thanks a lot for this, what preset would you recommend to isolate dialogue and music? we had a musical show and the kids where talking on the music very often on stage, i would like to isolate their speaking voice captured on video camera and afterwards put again the original music in Logic pro to make a gift video from the show for the parents.Thanks a lot for your answer
I would use a little bit of music rebalance and a little dialogue isolate. You don’t be able to completely remove the music but you will be able to reduce it
I actually have RX Standard (tried RX Advanced and it didn't get the sound out as well as Acoustica Premium). The combination of RX Standard and Acon Acoustica Premium Edition works extremely well. For what one misses, the other gets. RX Advanced just can't do it alone.
I'm just starting to delve into this kind of work and my biggest question is always: just how much noise is acceptable for a project? I tend to make myself go crazy trying to get every little bit that I can out, and I feel like I'm always balancing this fine line asking myself if it's better to take out more and subtly affect the sound of the voice or to leave it in and just accept it. Either way I feel like the result just sounds bad, though I've started paying close attention to audio in interviews, tv, etc. and I'm often surprised by the amount of noise left in the audio...
I would like to apologize for my previous response. I mistakenly mixed up your response with someone else's response on another video while communicating with me at the same time.
What would the proper order of operations be between this denoising part of dialogue editing, and "fill-and-fade" -ing with room tone? I have a commercial project I'm working on right now (I'm a picture editor, but I'm currently learning post audio and I'm doing the audio on this project) that has a scene that sounds exactly like your example in this video. But to make matters more complicated, there are two characters I'm cutting between and the intensity of the tonal noise varies between the different mics, as well as plenty of time when neither character talks. I started with denoising each character's lines, but then in the gaps between the lines, I didn't know what to do with the room tone. I have a nice BG under the scene as well, but it's completely drowned out by the tonal noise. And if I don't add the room tone in between the dialogue gaps, it sounds pretty jarring whenever each line starts and ends, even with the BG and the music underneath. I feel like I just don't understand what the correct order of operations of the dialogue editing process is. I'm gonna go back through and rewatch your first 4 "Audio Post Production 101" videos and hopefully find some answers there, but in the meantime if you are able to reply to this comment with some advice, I would of course greatly appreciate it!
Fill and fade first, make it smooth with long fades between each angle. You can find fill in the handles usually. You might need to piece it together from multiple tiny pieces. Then you can reduce the noise. In cases without much fill, you will need to generate it using Ambience Match (ambience only) , Undertone, or Absentia.
Learned more in this vid than a semester long sound design course. Thank you
These tutorials are such a valuable tool to anyone looking for advice and help in Audio Post. Thank you for making them
Thanks for this tutorial, we've just started dabling in RX10 Advanced from RX7 Elements, we found this tutorial really helpful and its opened up a whole new learning curve, but very helpful thank you
I'm dialogue editing right now and am dealing with this exact issue. Thank you for the tutorial! Perfect timing!
You also record with a generator running in your studio? 😬
@@1rcproductions I do film, there are generators lol
@@justkallmekai What are the generators powering?
I've embarassed myself today trying to get a gig as a VA Mastering dude for a Mod (No money involved), they couldn't take me because I didn't know what I was doing with Dialogue, since I've been learning how to work with music, not dialogue for films or videogames, this is a good insight into what is possible with RX, and I feel like with all that content out there, there are just a few that really explain how this software works, like this video. I'll do my best at getting the gig for my own good, I need that confidence for my future.
Great tutorial, thank you! I'd love to see a tutorial on how you clean nasty lav rustles and scratches 😃
you’re such a great teacher bro! than you
To clean up that type of filtered broadband noise I use Spectral DeNoise in a specific way (that's actually the only way I use that module). After you hit learn you'll notice that the range of that noise sticks out in the Noise Profile curve. Make sure Reduction Curve is enabled, grab that curve (the blue line) and drag it all the way up for the entire frequency range. Now add anchor points on the blue line just outside that filtered noise that's poking out and drag down enough points in between so that the Residual Noise curve (in yellow) looks smoother. Set your Reduction slider to taste and hit process. Give it a try, that usually does the trick. 😉
Yes that would work well. I just don’t like the sound of spectral denoise, always seems to leave behind some artifacts even on conservative settings.
Thanks thomas. Would love to see a future video on removing wind noise if had some time. Thanks!
great idea
Great tutorial. I'll have a chance to put this into practice this next week for dialog editing.
I had always learned that numerous processes being done to a single recording degrades it. It sounds like you have gotten better results doing several small processes...
Great video! Thank you, Thomas! Regarding broadband noise, I've found some interesting results with a very (-2/-3db) DX isolate pass on everything and then layer DX isolate for above 700-1khz and then Deconstruct below 300-700hz. Seems isolate retains the airiness of the dialogue whereas deconstruct will think it's noise. Deconstruct down below seems to find that noise between the harmonics better whereas isolate will not. For Deconstruct I'll do 0db tonal, -3/-6 noise, and opt to keep transient separate. Then leave sliders midway. Can work well for noise that often isolate and spectral denoise leave behind.
This is really good advice and something I started doing recently as well
this tool is great...I am new to audio editing...one of the challenges I have is whether or not to keep or remove noise...the dialogue becomes clearer however it sounds less natural...I was recording a speaker outside during a protest...letting the noise-people bleed into the dialogue gives it character even though the odd word here and there is harder to hear....
Keep the noise
The lower frequency hum I'd surgically remove because it's so prominent. Highlight the quiet area a la 1:41, pull up the spectrogram and move your mouse over each frequency (look for the obvious humps), go back to De-hum, choose static like you had in 2:02, but drop the the harmonics to 1 (not 16) to give you the ability to surgically removed each hum frequency. For lower frequencies, widen Q, e.g., 200-500. Rinse/repeat for the other lower frequencies. For the remainder, e.g., above 500Hz, just use spectral denoise, settings C or D.
That would be my approach too.
Great presentation, thank you for sharing your experience.
Beautiful tutorial. I will follow your channel. Thank you for the insight and taking the time on this :)
Glad it was helpful!
excellent tutorial. I'd like to learn this tool so keep em coming!
Very informative and insightful as always, thank you!
This is a great Tutorial, cheers.
Epic video please more of fixing on the go and saying loud why u choose that or another way
the noise level removel is incredible.
Extremely useful tutorial! Thank you.
Great tutorial,thank you!
Great instructional! But do you have a video with tips for getting the best “down & dirty “ results? My problem is dealing with tight deadlines & panicky clients so I seldom have the time to do such a beautifully detailed job like your example.
I try to stay away from jobs like that.
@@ThomasBoykin haha. But those advertising jobs are the best paying. You just have to put up with the clients.
Thanks so much, great work as usual
Thanks for the vid !! I was wondering how much time would it take you to do it like normaly, whitout having to explain while doing
A few seconds
Long time Tom, As usual cheers.
cool Im still learning Izotope can you remove the wind noise
Just subscribed to your channel-thank you. What school or courses you recommend for audio editing, professionally?
There isn’t one I would recommend unfortunately.
Any tips on how to remove mic handling noise of a recording of ambience?
High pass or very careful editing
This small steps approach yields impressive resultats indeed. As you mentioned, I'm one of those users going way harder on knobs, and not always happy witht the results. Intuitively I would have thought that with your approach, artefacts created at each small steps would add up and interfere with each other, even more distorted going through the following step, so the end result would not be good, at least not better than a 1 time carefully set pass. Seems it isn't the case. Thanks for the tutorial. One question : Could you polish the end result, which sounds a bit mids heavy maybe (be it in volume or in harmonic content) to me, by either EQing a bit, or use the spectral repair ? Thank you for this further info
For sure, but an EQ would be part of mixing, not dialogue editing
Freaking love these videos.
Hi, Do you have a video on how you would remove a cell phone ringtone ( Or anything similar ) from something like interview audio . I have RX 10 Standard. Not sure how much that would call for a different workflow in this case. Thanks
Yes, which ringtone?
It was the default Motorola "Hello Moto" musical type ringtone. So the ringtone intro is different from the body . It's not just a recurrent tone. . I hope I explained that properly. Not sure what the general term would be.
Thanks for taking your time to do those videos. they are really helpful for new and experienced people in post production sound. You mention resonance frequencies, do you find it helpful to use the eq to remove those after Noise reduction? I ask because removing resonance with EQ can be time consuming. What are your thoughts?
Yes I would remove them during the mix, or better yet redo the processing to keep them out of the final result
Nice video. The frenetic mouse movements make it a little hard to watch but I learned some good tips. Why did you decide against spectral de-noise? Using Learn on that seems like it would have done a good job with the background noise.
I save broadband noise for the mix since it is impossible to make those decisions out of context
i love your voice lol, dude i hate it how i like it lollllllllllllllllll. Imagine just a random stranger on earth, you being all serious, straight to point, and me just nodding and being like ah such a pleasure to listen to lollllllllll (no homo even if im a girl, just wanted to be honest, dont mean anything by it haha)
Have you tried Waves Clarity VX Pro yet? On some things it reduces a whole RX session into 1 knob. On others it annihilates breaths and anything not distinctly dialogue. Been using it since Thanksgiving on a feature and its potential is huge. Had a lot of noisy birds in a scene and it did miles better than Absentia DX's bird reducer. I'm curious to see how iZotope responds to Clarity.
Yes and I have other tools I prefer
Thanks for making more fantastic content
hello, thanks a lot for this, what preset would you recommend to isolate dialogue and music? we had a musical show and the kids where talking on the music very often on stage, i would like to isolate their speaking voice captured on video camera and afterwards put again the original music in Logic pro to make a gift video from the show for the parents.Thanks a lot for your answer
I would use a little bit of music rebalance and a little dialogue isolate. You don’t be able to completely remove the music but you will be able to reduce it
sorry forgot to reply, thank you very much! @@ThomasBoykin
Hi Thomas, what was the shortcut key you were using to go back and forth from the original and edited version?
Any suggestions on reducing hand brushing noise from a bodhran recording?
Try to use spectral rebalance and focus more on tonal sound
Nice vid man!
I would definitively have applied voice de-noise at the end, learning the background noise and gently remove it.
that's magic
You should take a look at Acon Acoustica Premium Edition, because there's things that it can do that RX Advanced can not.
I actually have RX Standard (tried RX Advanced and it didn't get the sound out as well as Acoustica Premium). The combination of RX Standard and Acon Acoustica Premium Edition works extremely well. For what one misses, the other gets. RX Advanced just can't do it alone.
I'm just starting to delve into this kind of work and my biggest question is always: just how much noise is acceptable for a project? I tend to make myself go crazy trying to get every little bit that I can out, and I feel like I'm always balancing this fine line asking myself if it's better to take out more and subtly affect the sound of the voice or to leave it in and just accept it. Either way I feel like the result just sounds bad, though I've started paying close attention to audio in interviews, tv, etc. and I'm often surprised by the amount of noise left in the audio...
Just take enough out so that you buy it as a natural recording of the location
Just take enough out so that you buy it as a natural recording of the location
Help me😢😢
How to distinguish just one person vocal from two people's conversations?❤
Spectral Layers 11
I don't even know how to here audio in rx its says no audio selected, help
I would like to apologize for my previous response. I mistakenly mixed up your response with someone else's response on another video while communicating with me at the same time.
No problem at all!
how can I use RX for free? the trial expired and I cant afford to pay that much :(
Time to pay up
@@ThomasBoykin -.-
What would the proper order of operations be between this denoising part of dialogue editing, and "fill-and-fade" -ing with room tone?
I have a commercial project I'm working on right now (I'm a picture editor, but I'm currently learning post audio and I'm doing the audio on this project) that has a scene that sounds exactly like your example in this video.
But to make matters more complicated, there are two characters I'm cutting between and the intensity of the tonal noise varies between the different mics, as well as plenty of time when neither character talks.
I started with denoising each character's lines, but then in the gaps between the lines, I didn't know what to do with the room tone. I have a nice BG under the scene as well, but it's completely drowned out by the tonal noise. And if I don't add the room tone in between the dialogue gaps, it sounds pretty jarring whenever each line starts and ends, even with the BG and the music underneath.
I feel like I just don't understand what the correct order of operations of the dialogue editing process is. I'm gonna go back through and rewatch your first 4 "Audio Post Production 101" videos and hopefully find some answers there, but in the meantime if you are able to reply to this comment with some advice, I would of course greatly appreciate it!
Fill and fade first, make it smooth with long fades between each angle. You can find fill in the handles usually. You might need to piece it together from multiple tiny pieces. Then you can reduce the noise. In cases without much fill, you will need to generate it using Ambience Match (ambience only) , Undertone, or Absentia.
I'm shocked when you say you can't hear the noise in some of your examples. What kind of audio monitors are you using?
just use music rebalance to isolate vocal
repair assistant do that faster and