Audio volume levels are a real mess

[AVITRY]

Hi all. Yet another new issue to deal with. I suppose these things are not really a big problem once I get a grasp of the easiest way to do things in PE3.

As I mentioned before, these videos are of professional pool matches. These are old S-VHS tapes I'm converting to DVD for a company. They were originally shot using two microphones on two separate audio channels. Channel volumes are often way off ( for example ambient sounds at the table like balls clicking an audience clapping are way too loud and commentators on the other channel are drown out).

Ultimately the company wants each audio channel to flow "normalized" and with the correct audio volumes relative to each other and then ..... they want the final muxing of audio to be mono on the DVD with both channels coming from both speakers.

Ok, that's the goal, but on this converted tape I'm running into a few problems I'm not sure of the best way to handle. The right channel, commentator speak, has various issues.

1. The entire channel is too loud

2. There are a few times the commentator moves or knocks into the microphone and it is ear piercing.

3. There are a few times when a good shot is made that the commentator gets excited and yells into the mic and this too is ear piercing.

I need to fix this entire channel, then have the volume output sound out correctly relative to the left channel before I mux it to mono.

My thought is to copy the audio timeline to another audio track then use the "Right Fill" effect on the copied audio (which will now be all commentary) to normalize it, keyframe out the knocking of the microphone, and perhaps even split the audio clips in areas that I simply want to mute the volume.

Then use the "Left Fill" effect on the original audio track to have both channels sound out the ambient ball and audio. And finally, ( and I'm not sure this is the best way) copy both of the audio tracks still again to two additional timelines and have 4 audio timelines and use the "Swap Channel" effect on the last two time lines, end up with all the sounds coming out of both channels when I finally encode it to MPEG.

I may be confusing myself with that last thought. without any further copying or channel swap effect would the two channels simply come out with both channels in both speakers?

I don't know, I'm really getting confused and concerned that all the mixing of effects will work correctly. When I use "Right Fill" and then use the keyframing and volume controls will it work on both channels since I'm doing it over the "Right Fill" effect? Will it play out on the timeline when I try it as I adjust it?

Does this sound like it will work and will the normalize effect get messed up if I try to keyframe real loud commentator screams or should I not use normalize at all and keyframe the extreme loud sounds and then gain up or down the clip as needed after I tone down the screams.

I'm just not sure the best way to proceed with PE3. I'm sure these questions seem silly to those in the know but I'm just not sure.

thanks in advance for any ideas!

joe

[Chuck Engels]

Hi Joe,
I would first keyframe the really loud parts, be sure to use a audio transition to minimize the change.
After you get it adjusted so the loud parts aren't ear shattering export the entire audio track.
Replace the old track that you adjusted with the new exported track and then Normalize that.

Have you worked with Audacity at all? It might do a better job of cleaning up the audio and since you need to export it anyway it might be worth giving it a shot.

[AVITRY]

Hi Chuck. Ya know I just think I've got a mental block when it comes to Audacity. I have downloaded it, and it really confuses me. I suppose tonight I'll just have to take the time to read through the entire help file.

thanks
joe

what do you mean by an audio transition is there a specific one you are referring to?

[Bob]

Copying the audio to another track and using right fill on one track and left fill on the other track will do what you want. That will place both the commentory and the ambient audio in both the right and left channels, so you will get mono with both channels coming from both speakers. Don't copy both tracks again and use swap channel on them, that isn't necessary and won't buy you anything. You can now adjust and keyframe each audio track individually as needed.

You won't need to export and reimport the audio unless you want to go further and do normalization or apply the Dynamics or other audio effects. Because you have multiple audio tracks, if you do need or want to do that, it is much easier to work with a single combined audio track.

Premiere Elements provides a cross fade transition for audio. It's like a dissolve transition for video only it fades out one audio clip while fading in the other. You don't use it with keyframes, only where you join clips.

[Chuck Engels]

Excellent advice Bob, thanks for filling in the blanks

[AVITRY]

Copying the audio to another track and using right fill on one track and left fill on the other track will do what you want. That will place both the commentory and the ambient audio in both the right and left channels, so you will get mono with both channels coming from both speakers. Don't copy both tracks again and use swap channel on them, that isn't necessary and won't buy you anything. You can now adjust and keyframe each audio track individually as needed.

You won't need to export and reimport the audio unless you want to go further and do normalization or apply the Dynamics or other audio effects. Because you have multiple audio tracks, if you do need or want to do that, it is much easier to work with a single combined audio track.

Premiere Elements provides a cross fade transition for audio. It's like a dissolve transition for video only it fades out one audio clip while fading in the other. You don't use it with keyframes, only where you join clips.

Great Bob, thanks! Now if I use the normalize effect for the commentary, and then keyframe the extreme loud parts (of vice versa) will the normalize effect try to even further mellow out the commentator's screams? I mean to some degree, I do want to be able to allow some emotion and excitement in their voice. I don't want normalize to completely flatten it.

Thanks again guys, I'm learning. Most everything I've had to do to these VHS films has been color correction and sharpness, and that's a no brainer with the Canopus 300. All the audio tweaks that I have been doing were just one thing. I've never tried to apply multiple effects over one another on these clips yet so I've got a lot to learn.

[Bob]

Keyframe first and then apply the normalize effect. Effects are rendered in the order, from top to bottom, that they appear in the effects properties panel. The built-in volume effect, which is what you will be keyframing, will be rendered first.

Rather than use normalize, try using the Dynamics effect. It's similar to normalize, but you have more control. The help file describes what the different components of the effect do, but the two you'll probably use will be the compressor and the limiter components. Limiter reduces the peaks while Compressor will compress the dynamic range. You'll mainly just need to enable the component and set the threshold values. You can adjust the thresholds and enable/disable the components while playing so you can hear what's happening. The autogate component may be useful if your vhs audio has a low level hiss, you might be able to reduce or eliminate some of that.

[AVITRY]

Thanks again Bob, that Dynamics Effect sounds like just the ticket!

[AVITRY]

Hey guys, I gave up on the Premiere Pro 2 audio and I'm back to using Elements to correct the audio on this capture.

I was expecting something different to happen I guess. I unlinked the audio and video,copied the audio to the second audio track, I applied the Fill Left effect to the first audio track and the Fill Right effect to the second track. ( the second audio track being the commentary, while the first is the ambient sounds like balls clicking)

I'm now ready to keyframe the second audio track where the commentator screams so very loud, but I was expecting to see the WAV graphic on the audio lines to be the same in each channel. They are not. They stayed graphically representing two separate channels with different wavs.

If I move the second track out of the way, the 1st audio track does only play ambient sounds and when I move it out of the way and bring back the 2nd audio track, it does only play commentary as they should.

Why do I not see a single syn wav or at least two duplicate wavs on either given track? Do I just keyframe using the half of the track that I know is being played or is there a way to change the display?

Dumb questions ... huh. I'm sorry. Just didn't make sense to me.

[Chuck Engels]

Hi Joe,
You have only added effects, you haven't really made any changes to the original audio at all.
If you export the audio of the entire project and then replace your existing audio with the new exported copy you will get what you are looking for, two channels with identical wav forms.

[AVITRY]

Thanks Chuck, I keyframed based on half the display and it seemed to work. Funny, I just assumed that if you were going to be able to hear the effect in real time you'd see it too. No matter, I guess, just another example of how preconceived assumptions on my part lead me to confusion when they don't end up with what I think is the logical result.

FWIW I did look for a way to render the audio on the timeline. I thought that might result in seeing what I'm hearing. I didn't find a way to do that either As Curly says from the 3 Stooges, " I'm trying to think, but nothing happens."

[Jayell]

As Curly says from the 3 Stooges, " I'm trying to think, but nothing happens."

I love that! It fits so many situations