Hi,
I’m wondering if there is a better way to fill ambience/roomtone in dialogues beside the old traditional way of looking for leftovers from what recorded on field.
I’m familiar with the Rx ambience feature , its not bad for matching ambiences but i looking for something that will create new file that contains only ambience based on a reference that was recorded on field.
I hope i’m clear 
Thanks,
Oren.
Rx ambience match can do that…
PS:-----
I realize now after writing all of the below that maybe you haven’t used RX the way I use it. Here’s my procedure:
Now I have a file with only the ambience match generated tone.
--------- here’s what I wrote first ---------
My thoughts on RX ambience match is that it varies a lot on a case-by-case basis.
First of all there’s a bug where low tones aren’t correctly picked up and recreated. So if you have a low hum of some sort it may register as a different pitch, making it unusable as fill.
Secondly, I find that the best way to get usable fill out of it is to grab those leftovers in handles and create a new file out of it. I then bring that into RX and de-click it and take out things that I think the algorithm can get confused by. So if there are some extraneous noises that are long or loud enough, especially if they contain a clear pitch, I use tools like spectral denoiser and declicker etc to do a quick and rough cleanup. Then I get a decent tone out of it about 7 times out of 8 or so.
Another thing I think matters is how this is to be used. I’ve been recently working quite a bit on a show that takes place near beaches so there’s always a fair amount of waves / water / wind involved. When there’s a lot of movement RX doesn’t work because it’s too static, but when it’s a more static environment it really helps edits because I can quickly create tone that by itself sounds nothing like the ocean (no movement), but really helps smoothing transitions between shots. In other words I’ll create the tone just so I have something clean that brings up that range of frequencies that exist in shot B smoothly, and also lets them trail off when moving to shot C. In and by themselves they don’t sound good, but because we’re cutting to a different shot with dialog at the top of it the ear doesn’t notice the static nature of it, so it helps a great deal (of course, now everything is more noisy during those transitions… but… what do you expect when shooting audio on a beach and then massacring the dialog in the edit…)
And then lastly I’d say that for more quiet and delicate content RX ambience match still works very well once you fill in the gaps. I actually feel quite confident in my ability to create very clean dialog tracks now when production / dialog editing, but whether or not it works in the end depends a lot on what else is added to the mix. If you add foley and other effects it generally works very well. If there’s nothing but speech and tone it doesn’t. And if that’s the case I find that using that generic RX ambience fill works if I also add just some ‘sounds’ from the set. These could be gentle movements (fabric, clanks etc) or whatever, but on separate tracks and actually cleaned up a bit. Gives me more control.
Other than that I don’t really know what a good technique would be.
I haven’t used rx6 with Nuendo yet…
But with PT I just select a clip that I want to amb match, hit learn on ambience match plug in, select output ambience only (tick the box).
Then I select (paint) the lenght I want/need on the track where I want the new amb match clip to appear and just hit then render.
Thanks.
i will sure give it a try.
Yup on PT this specific feature is much easier to use, unfortunately in Nuendo you must select a part with audio and not just a region on your timeline.
That sucks
It’s not an answer to your question, but pass a stern message back up to the location recordist to make atmos / roomtone recordings ( 1 minute) for each location.
Perhaps for each camera angle if that makes a difference …
‘Traditionally’ you’d be fired for not doing this … the fact that having to dig around between lines for ambience has become a ‘tradition’ in itself is endlessly frustrating and needlessly time consuming.
ltf3,
That’s going to depend on the market and type of work though. For me doing mostly non-scripted TV here in NY there’s never any time for us audio engineers to hunt for tone on source media, which means that we use whatever the editors decide to put into the sequence and then export via the AAF/OMF. Unfortunately they have less than great listening environments and very little time, so 8 times out of 10 the “tone” I’m given isn’t really the tone that’s needed to fill dialog tracks. It’s either a looped section with junk in it that the editor didn’t hear in his room, or tone from a different angle/take in the same room, meaning at that point the air conditioner was running at only half speed or the truck outside was idling or whatever.
Nobody gets fired for screwing that up unless we’re talking bigger networks, and even then a lot of the time those in control of the shoots care mostly about how it looks and if people say what they need them to say (or close enough). Audio is then “fixed in the mix”… 
For film and higher budget scripted TV though I absolutely agree.
Can you make a “dummy” event and render on that?
Bye /Tumppi
I haven’t seen proper room tone recorded for a long time now. Features and television drama.
I’m just happy if the dialog sounds good, and although having proper RT is great creating some useful fill is often faster than searching for it anyway.
I have never seen RT recorded specifically for all angles. But if I can get just a few quiet seconds between sound rolling and action I’m happy.
Even “Quiet on the set!” seems to be gone these days. I now repeatedly get not only fill with the crew yapping in the background, but even edited takes of “talent” with some dumb crew person talking in the background…
I have a feature next year where last week the sound recordist apologised that he was not able to get camera crew to respect silence during takes. Steadycam used extensively and DOP and steady operator was continually talking to each other during shooting dialog so a huge amount of ADR expected.
So god damn stupid.
Well, as long as the production sends their money over to us to do ADR I’m all for it. More money!
The problem really is when they don’t respect sound in production and then just expect us to fix it in the mix, because, you know… no money…
(which btw means the producer still wants to buy a new car and doesn’t want to sacrifice that just for stuff to sound better)
I’ve gotten a fair amount better at cutting out background talk out of desired dialog using RX. But oh boy is it tedious and literally thankless work…
Working extensively on Dialog. 95% of the sound get processed in RX from Nuendo. Here is my 1 shortcut roundtrip between Nuendo and Rx :
Some advices :
With the ambient match feature (remember to click the « ambience only » option), I don’t need any room tone anymore.
IMHO RX fill is great but oh so utterly static and boring so yes it is extremely helpful but of course it isn’t the same thing at all.
But it sure beats finding 8 frames and looping it…
Nope 
That is just stupid