How to merge a composite shot with separate audio

djfrododjfrodo Website User Posts: 68

Before I begin I'm going to apologize for the barrage of future questions I'm about to unleash, not only in this but other threads.

Here's why. I shot a short film, learned some cool stuff, made a rough cut, and now I'm trying to put it all together.

So, here's my latest question and my process.

1) I filmed a scene when someone gets shot

2) I merged the audio from an external recorder with #1 - everything went great

3) I did tons of test and learned how to replace the sky in the shot, and it looks great - but I used the raw audio (#1, not #2), which was dumb, but I was learning visuals instead of sound.

4) I learned how to animate bullet casings and added sound effects from  #3.

The problem is the merged audio from #2 isn't used in #3 and #4.

My question is twofold - the first is about editing in general.

Is it a better idea to get all sound merged and exported before playing with visuals (I'm thinking yes)?

Second, is there an easy way to merge a composite shot with separate audio (as in merging the audio file from #2 and the visuals from #4)?

If I try merging the last good visual scene (#4 - so bullets and sky replacement) with the separate audio I don't even get a menu option to do so.

I really hope this makes sense, but I can also offer any newbies advice - when people tell you to organize your workflow, do it. When you're learning it's fine to just learn, but the first time you want to take everything you've learned and stick it together you better have organized your media, or you're in for a hell of a ride.

Comments

  • Triem23Triem23 Moderator Moderator, Website User, Ambassador, Imerge Beta Tester, HitFilm Beta Tester Posts: 18,207 Ambassador

    Sound is AFTER visuals. (Anecdote: I was audio designer for an internet media company. They insisted on having a full audio mix before animation. I still have my memos discussing how utterly stupid that was and guaranteeing audio would have to be redone, putting us over time and budget. When the CEO asked why we were over budget I gleefully sent the entire thread back with a pointed suggestion he listen to me next time.)

    Sound is ALWAYS after visuals. Mix dialog first, then SFX, then background ambience, then music, then final adjustments.

    In general Composite Shots aren't where you want to be doing your audio editing. Once your effects shots are done, they should end up on the Editor Timeline with the rest of your film. The Editor Timeline is where you have the full audio mixer, etc. Just layer up extra tracks.

    If your audio from take 2 is the same as the external recorder you can merge them in Hitfilm's normal way (right-click both in media panel, select Merge Audio--it could be Synch. I'm at work and doing this from memory).

    As far as merging two completely different audio tracks, you could bury them both in a Composite Shot (the Composite Audio will become a single track in the Editor), but, again, this is bad workflow.

    You really don't want to start "merging" audio clips together. Just mix audio AFTER your picture is locked. Don't be afraid of multiple tracks. Layering is good. 

  • Triem23Triem23 Moderator Moderator, Website User, Ambassador, Imerge Beta Tester, HitFilm Beta Tester Posts: 18,207 Ambassador
  • djfrododjfrodo Website User Posts: 68
    edited March 2018

    Thanks for the reply.

    re: "Sound is ALWAYS after visuals. Mix dialog first, then SFX, then background ambience, then music, then final adjustments."

    Dialogue is sound...and I think I might have not stated my use case very well.

    I merged an audio and video track...and it was perfect (Hitfilm is actually good at this).

    Then I did the exact same scene with a sky replacement, and the bullets, but with the scratch audio.

    The audio here is diegetic only.

    Foley obviously comes after the dialogue and visuals are locked.

    What happened to me is I did the sound first, then forgot to use that version and learned sky replace, visual fx, etc. and it looked great, but the sound SUCKED (constant hum of the camera trying to cool itself).

    So...I think you may be right and that I'm probably saying the same thing. Dialogue (not "sound") first, then visuals, then "sound" (foley, music, etc.).

    Doing it the other way seems to me it would lead to major problems. Any small edits in the visuals would lead to much more work to sync the dialogue.

    If the raw video is already synced to the diegetic sound and exported, then you can do whatever you want with fx and foley.

    The other way just seems backwards to me.

    But, I could be wrong.

    p.s. I tried what you suggested about an hour ago and the video and sound are off by a frame or two, and it's just not working.

  • Triem23Triem23 Moderator Moderator, Website User, Ambassador, Imerge Beta Tester, HitFilm Beta Tester Posts: 18,207 Ambassador

    Yeah, correct. Dialog will come first, but even then I'd use the production audio first, and worry about cleaning things up or dubbing later in the mix.

    If audio synch is drifting, check to make certain your source footage isn't variable frame rate. Hitfilm bravely tries to correctly parse VFR but it's problematic. 

    About 19:30 in this video using MediaInfo to check for VFR is covered. The latter portion of the video covers transcoding. 

    If your video is VFR, OK, mildly annoying. But, once you transcode the footage, right click existing footage in the media bin, select Relink then select the transcoded file. This replaces the original file with the transcode but preserves your edits.

    Different editors might give slightly different answers, but, yes, synching an external recorder to camera audio can be smart. I tend to wait, because lets say I have five takes. I'll use one. Why take the time to synch four takes I won't use?

    Another edit trick. Let's assume you've synched audio later in the edit. I'm 90% certain this synched version replaces the original on the timeline. If I'm wrong and it creates a new media panel entry... Ok, holding Alt and dragging the new clip over the original replaces the original with the new version and preserves the edit. 

  • djfrododjfrodo Website User Posts: 68

    re: Yeah, correct. Dialog will come first, but even then I'd use the production audio first, and worry about cleaning things up or dubbing later in the mix.

    Very interesting. This is my first shoot, and I've just run into the problem of either sound being great and the footage sort of being meh...

    Or, through the wonders of fx, the footage looking great but then syncing sound being terrible.

    Basically I'm just trying to avoid the 70s kung fu sync while still keeping the vfx.

    Btw, all of the footage I'm dealing here is not vbr and was shot on a nice camera, so I think it's just a matter of taking the time to edit/composite the visuals by themselves, then match audio, or sync audio first then add fx.

    Doing both separately and then trying to merge them is problematic. 

    I have noticed that more than about 3 seconds between production and external audio throws off the merge/sync...

    Thanks for your help. I'm sure I'll keep doing this until I get it right.

     

     

  • AlexofAllTradesAlexofAllTrades Website User Posts: 5

    What do you do if your audio is the start point of your video? Podcast/VO is the start, video is mixed based on the audio. Doesn't it make sense that I should be able to hear the audio in the composite shot then? Not all videos (dependent on genre) would audio be last. Think of Yahtzee style videos. His script and the rate at which he speaks and what he happens to be saying informs the visuals, so in cases like that audio is first.

  • CleverTaglineCleverTagline Moderator Las Vegas, NVModerator, Website User, Ambassador, HitFilm Beta Tester Posts: 3,025 Ambassador

    @AlexofAllTrades You are correct. In situations where there is no image captured up front, and the pace of the project is set by a voiceover track, then that's where you start. However, it's still best to hold off on mixing in other audio — sound effects, music — until after the visuals are completed to match the VO. Most of the discussion above your comment is related to traditional filmmaking, where you're capturing specific visuals of speaking characters as part of a story, which is why @Triem23 emphasized focusing on picture first.

This discussion has been closed.