Question about dialog editing.

I

ironpony

Carbonite Member
I have been trying some ADR for the first time, and having trouble. When I match the syllables from the audio to the syllables from the video, the sound makes static noises when going from one syllable to the next, after I have put them together. How do I get rid of the static, or what am I doing wrong? Thanks.
 
1) What do you mean by "static"?

2) Can you post a sample so we can actually hear what you're hearing?

3) How are you matching syllables? Are you just chopping and sliding, or are you using an ADR plug-in such as VocAlign?
 
Very interesting. VocAlign does something similar but you can't use it on a clip that long and have it work well.
 
Alex H. said:
1) What do you mean by "static"?

2) Can you post a sample so we can actually hear what you're hearing?

3) How are you matching syllables? Are you just chopping and sliding, or are you using an ADR plug-in such as VocAlign?
Click to expand...

When splicing and dicing audio, shouldn't there be some small amount of fading crossing from one chunk to the next. I've not tried 'syllables', just whole words, so there were small 'pops' that fading to room tone covered. Well, perhaps 'pop' is too strong... 'bumps'... between cuts.

Or are syllables so 'small' that no fading would work.
 
more than likely the clicking or spits as I call them are happening from a wave for disparity or gap, in other words if youre splitting up or creating one track from several takes and manipulating their place in time the wave form from the previous segment must match the wave form place in amplitude of the following segment or you will hear a click because youre literally re-creating what digital noise does by accident.

to prevent this you can set your workstation to automatically crossfade the segments by their overlap and this should smooth out the coincidental wave form.

be wary that if your edited segment is too small and doesnt have a small tail to crossfade with you might still get this artifact because there will be an amplitude gap between the segments.

Hope that helps.

GG
 
ironpony said:
I have been trying some ADR for the first time, and having trouble. When I match the syllables from the audio to the syllables from the video, the sound makes static noises when going from one syllable to the next, after I have put them together. How do I get rid of the static, or what am I doing wrong? Thanks.
Click to expand...

As I read your post - if you're listening in mono with the studio audio on top of the video audio, that might cause your static .. it could be a phasing problem.

It's good to trial match it that way, but listening to your 'new' voice tracks with the video is how your audience will view it. The one thing in your favour is, they won't be hanging on every syllable and sorry, but you won't get it absolutely perfect.

So once you get your possible OK take, play everything from 15secs back to see if it'll pass muster.

And if you're using any new actors to this field, watch out they don't slowly lose it and fade away .. keep a ear on that. Good talent at ADR can be upset by a newbie having trouble.

Keep as many re-takes as possible in case that happens, and in case enthusiastic talent keeps saying 'but I can do better than that' and ends up not doing it.
A 5 minute rest can work wonders.

And sometimes it's good to start work near the end so new talent can settle in, then work from the head of the show.

It takes some experience to juggle all the factors involved in producing professional ADR every time .. hang in there mate :)

Cheers.
 
Ironpony doesn't seem to have returned to the discussion he started...
My first guess (in the absence of any useful details) was that "match the syllables" means some sort of manipulation of the length of bits of audio track (by an unknown method with undisclosed software). Of course some NLEs are better than others at preserving the integrity of the audio track under this kind of manipulation. We don't know whether ironpony is using adequate tools or even using a basic premise that is valid. Or even whether he is working with talent that just can't reproduce their lines well enough to make ADR even practical in this case? Way too many unknowns to allow for anything but wild shots in the dark.
 
I agree with Richard. There are too many unknowns for me to even begin to know where to start. I need to see/hear the audio. And by see I mean take a look at the audio session (in whatever DAW is being used) to see how it is actually cut.
 
Sure I will upload a track. It's not the ADR talent behind it. In the original recordings, the acting is actually good. But when I cut and match up the syllables there is popping sounds in between as someone put it. I am not using vocalign, and just putting the syllables together manually. What do you mean when you say you seen to see the 'audio session'?
 
ironpony said:
In the original recordings, the acting is actually good. But when I cut and match up the syllables there is popping sounds in between as someone put it. I am not using vocalign, and just putting the syllables together manually.
Click to expand...

So you're literally just cutting around words and syllables, and sliding them on the timeline to match the actors' mouth movements. This means that a) when you separate words and syllables, there's a gap left behind which has none of the room tone behind the dialog and; b) you're most likely cutting the waveforms in places that actually create pops in the audio. If you move words or syllables closer together, there's no gap but there still may be a pop. Pops can be tamed with slight fades at the heads and tails, or slight cross fades where regions overlap, and the gaps can be filled in with room tone. But this isn't going to give you results you'll like. Things won't flow naturally, and the filled-in gaps will sound like just that: filled-in gaps.

This is not the way to edit ADR tracks.

What VocAlign does is utilize a variable rate of time compression/expansion to stretch or compress words and syllables so they not only start in the right spots but also lose nothing in between (leaving no gaps and creating no pops).

ironpony said:
What do you mean when you say you seen to see the 'audio session'?
Click to expand...

I think Dave is talking about a screen shot, though I'm not sure that's necessary since you've clarified your workflow. Are you doing this in a DAW such as ProTools or SoundTrack Pro, or are you trying to do this within your NLE (Final Cut, Premiere, Vegas)?
 
With out hearing it but based on your description, you are getting pops on your edits. That will happen unless you cross fade them. Some apps have a "zero crossing" preference that should also work. If you zoom way way in you will see at your edit points that the waveform doesn't line up. Anytime you have a waveform that doesn't line up on the cut it will pop unless you cross fade.


As a more general note I don't think you are going to get decent results cutting the way you are. You may be "in sync" but it's going to sound real choppy.
 
ironpony said:
Sure I will upload a track. It's not the ADR talent behind it. In the original recordings, the acting is actually good. But when I cut and match up the syllables there is popping sounds in between as someone put it. I am not using vocalign, and just putting the syllables together manually. What do you mean when you say you seen to see the 'audio session'?
Click to expand...
A great deal hinges on exactly what YOU MEAN by "cut and match up the syllables". Until you precisely and completely define what you are doing (and what tools you are using), we will just be dancing around in the dark here. A waste of time for us, and of no benefit to you. If you don't provide the required details, we can't help you.

I don't see any benefit in seeing/hearing your original footage. The issue appears to be your METHODOLOGY for trying to fix it.
 
It sounds like if it is popping sounds you are not cross fading and/or cutting on the zero point (although you can cheat that if you absolutely have to). I've seen really good dialogue editors work magic. The actor Jet Li did a movie called "Rome Must Die" some years ago. His english was not very good at the time, and when the movie came out, everyone remarked how much is english had improved. What they didn't know was the dialogue editor actually cut in consonant sounds we wasn't pronouncing into his dialogue track so seamlessly you couldn't tell. That's when I learned dialogue editing is an artform.
 
I tried uploading the track to sound cloud but it will not accept it. I cannot use vocalign cause I do not have pro tools or anything that is compatible with it. I have the Adobe CS 5.5 creative suite. I tried the different types of fades in the crossfade file but none of them worked and there is still popping. Thanks for the input so far though! But I think I will need a step by step instruction of what I am suppose to do in either Premiere or Audition. Thanks.
 
Last edited:
ironpony said:
I tried uploading the track to sound cloud but it will not accept it. I cannot use vocalign cause I do not have pro tools or anything that is compatible with it. I have the Adobe CS 5.5 creative suite. I tried the different types of fades in the crossfade file but none of them worked and there is still popping. Thanks for the input so far though! But I think I will need a step by step instruction of what I am suppose to do in either Premiere or Audition. Thanks.
Click to expand...

I'm almost positive there's a crossfade option in Audition...

Try this link.
 
You can get rid of those pops by using a cross-fade (as short as possible) between the two audio tracks at the offending cut. You can do this very easily in Premiere.
 
Thanks but I tried in Premiere but when you got crossfade, there are three options. Gain, power, and exponential. None of them seem to work, unless I did it wrong.
 
You must log in or register to reply here.
Back
Top