logo Sign In

Info: editing (blending, syncing, adding/subtracting) audio

Author
Time
 (Edited)

I’ve used NLEs before for short, basic projects that I’ve shot myself, where I control all the original audio and video input sources. I’m competent with the basics of how NLEs work. But I’m utterly mystified as to how the better faneditors will shorten or lengthen a scene and have the music and atmospheric sounds flow seamlessly. I imagine it only appears to be seamless, but music-wise, it has to be pretty much perfect. In terms of atmospheric sounds, I’m sure there can be more leeway. And dialogue starts and stops completely enough that I’m not at all confused about how that would be edited, as mentioned above, I have edited before.

So really my main question is about music. When the original scene lasts from 1:00 to 2:00, and has music beginning at 1:00 and ending at 2:00, but I want to extend the scene to last from 1:00 to 3:30, having the cues for the swells or crashes in the same places relative to the visual events, which are now at different times, how do I do that? I’ve thought about it and it seems that without tracks being separated (sorry, an unspoken assumption here) moving music around would be impossible.

Are a lot of editors finding cleanly separated audio channel sources or something? I’m assuming at best faneditors get a video track and a right, left, and center audio track, all of which have the same audio but in different volumes.

So really my main question as about how you edit sound when the source has all the sounds mixed together in a single track, or several tracks each of which have a mixture music/atmospherics/dialogue 😛

Do you use filters to artificially separate the three elements (foleyed, dialogue, music)?

Anyway, information anyone? Thanks!

Author
Time

I’ll attempt to answer a few of your questions as best as I can.

Depending on whether you’re editing at Stereo movie or a 5.1 surround movie (and also the method with which you ripped that particular audio) You’ll either have two tracks, left and right, or 6 tracks, left, right, center, surround left, surround right, and Low Front End (LFE).

In stereo, it’s almost impossible to isolate a particular part of the audio landscape. In 5.1 it’s very difficult, and not always possible, but you have a bit more leeway.

There are some videos that have completely isolated center channels that only contain dialogue and maybe a few sound effects here and there, but will have no music. The Clone Wars series is a good example. For instance, you’d be able to completely re-score those episodes by just replacing the music tracks and maybe adding a little bit of Foley work to fill in the surround channels.

Most movies, however, will have muddied center channels, that contain dialogue, music and special effects. Most of the time the music is quieter on this channel than the others, so removing the other channels will get most of the music out. I’ve had some success by doing this and playing with the levels of the videos so that the old music track is barely noticeable.

Star Wars is the most vexing of films because it’s center channel is all over the place, hence why very few of them have drastic audio edits.

I have seen mention of programs that can separate these sounds further, but they are imperfect by nature, and I don’t have any experience with them. Perhaps someone else can give you some more info on that front?

I hope this helps, and good luck on your project!

Author
Time
 (Edited)

@SpenceEdit: That paints a lot into the picture for me, thanks.
@Hal 9000: Thanks for confirming that a lot of editing is just being convincing in what are probably ridiculous sound hacks ^_^

So it sounds like a lot of the effort is spent using hacks to get music out of your way, presumably so that you can omit it alltogether or add it back in, synchronized with whatever scene you’ve inserted. Is this correct?

An obvious follow up question is, what do you do where you cannot get the music out of the way/replace it? Does the music then dictate what you can or can’t convincingly do, edit-wise?

Author
Time
 (Edited)

Ouch. That was the one possibility I was hoping would NOT be an answer. Wow. So entire edits are dictated by certain scenes where music and visuals are crucially linked in the originals. Okay.

THAT is enlightening. I figured maybe some editors were either good enough or diligent enough or clever enough to get around that somehow.

Good to be aware of the immovable objects at least!

Author
Time

I’m only responding here because I was pointed here. In fact most of my work is far from invisible, unfortunately there is not much I can do about it given my sources. One thing I have dabbled with is Spectralayers, which offers a way to “paint” out portions of the sound visually while retaining others. It seems pretty useful, but tedious…and expensive.

Author
Time

This is where the editor’s creativity comes into play. There’s a section in my Hobbit edit where I combined three different scenes into one sequence (The bit where Thorin gets his sword back from Legolas in the battle, if you’re curious) and it took a ton of tweaking to get the audio right.

Is it completely invisible? No.
Is it perfect? No.
If the edit is working in a story driven, emotional way and you’re not looking for the change, does it work without being too noticeable? Absolutely.

That’s the best way to look at what you’re trying to accomplish. 90% of the time, what you want to do is possible, you just have to get real creative with it.

Author
Time
 (Edited)

The old standby of those who create: ITS ALL A HACK.

^_^

Spectralayers sounded interesting, but I came across something brilliant while looking up alternatives and similar options. This article: http://www.howtogeek.com/61250/how-to-isolate-and-save-vocals-from-music-tracks-using-audacity/ suggests that to isolate vocals from music you need the original vocal-less soundtrack, which you then basically invert and superimpose over the version of the song with vocals. Here, what is not in the instrumental, is by definition the vocals, and as long as the two instrumental tracks are indeed the same track, apparently with some tweaking Audacity can deduce a new audio track (the vocals) from that difference.

In songs, this technique would rarely be useful, as most aren’t released in a vocal-less version (although maybe artists officially distribute a karaoke version, idk) but in the case of movies, any movie that releases its soundtrack (which is common) should be able to have the in-film dialogue isolated/extracted, at least in theory. I wonder what the quality would be though. If you’re mixing music back into the finished audio, any distortion might at least be masked by that.

I’m also guessing scenes where explosions and lasers and whatever are present in addition to the music would probably pollute the ideal vocal result. But again, we’re kind of talking about hacks in the first place, so maybe that is tolerable.

Author
Time
 (Edited)

Meaning in order to isolate the dialogue (into its own separate track) in a section of the original audio where the only significant sound obscuring the dialogue is the music, you simply need the original score. Which is widely available. It sounds just like working with layers and transparencies in an image editor.

Author
Time

I know this is serious necroposting, but I am only just recently really getting my feet back into the editing community. I just wanted to note that the Audacity avenue seems to be the holy grail, or unicorn of audio editing. Many have tried, but rarely have any achieved any success. I believe Hebrides over at fanedit.org successfully replicated that process for one scene, and the results were simply astounding… but even with his detailed instructions I don’t know anyone that achieved success again, and even he noted that it only worked that one scene/time… using his same instructions he himself never replicated it again on other scenes (that I’m aware of, again, it’s been a while).

At the end of the day, most faneditors have to use the music within the scene and work around it, find interesting ‘hacks’ as you say to make it feel natural, and that’s the key, there are many tricks manipulating exactly when the tracks overlap and crossfade (moving them before or after the visual edit) that can help mask the edit, but fanediting is largely something you do with one hand tied behind your back.

njvc uses a combination of spectral layers (as spence noted) and sheir blunt force to achieve fairly impressive audio replacement. He uses spectral layers to remove as much music as possible, and then he tends to overpower whats left with strong audio replacement. Sometimes it works brilliantly, sometimes it’s just enough to make it acceptable, but you can still hear it if you listen.

I hope your project turned out, or is turning out well 😛