Pumping Up the Volume

solarblast1 wrote on 12/10/2008, 5:06 PM
I've been looking at a large collection of audio and video event I've produced for a video that make the whole. I've particularly been working over the audio portions with plug-ins, envelopes, ... I decided to Render As an avi with NTSC format to square it all away. The resulting audio track surprised me. The audio volume seems "smaller" visually than before. I can pump it up with the mixer fader, but it doesn't quite seem to be so loud as in the unfinished state. What, if any, plug-in can I use to pump it up further without distortion? Another shot with Hammer on the now single audio event? Why does the volume seem to be transformed to something lower than before?

Comments

newhope wrote on 12/10/2008, 7:33 PM
All compressor/limiter plug-ins, of which Wave Hammer is one, actually reduce the dynamic range of your audio. To compensate for this they then allow you to shift the overall volume up. Depending on the actual plug-in the method of doing this can either be automatic or manual. It can also be set off.

If you have set the autogain off then the dynamic range is reduced by the compression ratio and threshold interrelationship but you have to offset for the reduced peak volume.

Have a look at the Compressor plug-in that is part of every track in Vegas. Turn it on and set the ratio to about 3:1 for speech then reduce the threshold until you see gain reduction starting to indicate on the gain reduction meter in the Plug-in (note the difference between the Input and Output meters of the plug in as well) Down the bottom left you will see a check box for automatic gain makeup. If ticked it will automatically make up for the overall gain reduction. If it is off you can manually set the gain increase. (I'm typing off a Mac at present without being able to see Vegas so can't be as accurate as I'd like with the actual labelling of the plug-in). I would personally use a combination of the track compressor (at around 3:1 compression ratio) on the individual tracks and then have Wave Hammer on the Master Fader for the type of speech based video I've seen you have uploaded.

So as you can see you need to either set up a manual gain amount in the compressor or leave the automatic gain make up on.

If you do this the average level of the audio will be louder because, by controlling the loudest peaks with compression/limiting then moving the resultant overall level higher you are also increasing the volume of the lower level audio and therefore the average volume of the track/clip.

This, if done correctly, shouldn't require a second level adjustment after you have rendered your video/audio track with Render as.... The final render should be correct.

New Hope Media
rraud wrote on 12/11/2008, 7:21 AM
With the channel, master, event and automation volume levels set @ 0, theoretically there should be no change in level. Unless there is a (known or unknown) audio plug-in process happening somewhere... or rendering to a lossy or other audio format could change it also.
solarblast1 wrote on 12/11/2008, 9:28 AM
I'm certainly a neophyte when it comes to the audio side, but have been on a campaign to understand what's going on. Bare with me. I have a reasonable understanding of Db, at least from a computational side, and understand the filter concept.

I've been working on this project long enough that I'm not sure I turned on the track plug-ins or they were defaulted on. I have become aware of them, in any case. All three are in play. In my case, gain (track, Compression, Amount) has a default value of 1. The plug-in help is clear on what that means. It is not clear on what threshold means, which is a whole other issue in Vegas land. However, an entry in Wikipedia is clear (Googled: audio compression threshold--first hit). The insightful words on threshold are mention of input and output, which the Vegas Help does not mention.

I think I need to read that entry carefully before going further. However, I'm sure your recommendations are correct. I'd just like to know what I'm really doing.

I'll mention the only change I made in Hammer was to change the volume threshold to -12.6, and applied that to all audio events. It seemed to put every event on a fairly even volume. After that, I applied Render As. Last night I rendered 9 minutes of the new version of my effort and played it on my DVD player. Not bad. It still may need a boost in volume. However, there may be need to be an improvement in voice clarity.

The video is made in an elementary school gym where 6 presenters are demonstrating physics concepts to 3-6th graders. I was able to get within 20-30' of the presenters most of the time with my camera, which has a built-in mic. It works pretty well, but there are moments when some audio clarity to a voice might be helpful. Nevertheless, I think it'll be fine for its purpose. It's certainly giving me a lesson in the audio side of video work.

Thanks for the response. It has proven helpful.
newhope wrote on 12/11/2008, 1:49 PM
the only change I made in Hammer was to change the volume threshold to -12.6

Setting the threshold at this level, as an example, will mean that whenever the audio level exceeds -12.6dB (from the 0db maximum - 0dBFS) Wave Hammer will apply gain reduction.

The same will apply to any Compressor or Limiter. Signal that exceeds the threshold will be gain reduced when it exceeds the threshold. The amount it is reduced is set by the compression ratio. Hence if you set a threshold of -10db and a compression ratio of 3:1 then for every 3db the audio level exceeds -10db the output will rise by only 1dB as gain reduction is applied. This makes the loudest peaks in your original recording quieter meaning you can raise the overall level making the whole recording louder without exceeding the maximum 0dBFS and producing distortion. This is where automatic and/or manual gain compensation comes into play by allowing you to increase the overall volume of the audio in the compressor plug-in after the gain reduction has been applied.

Of course care needs to be taken because over compression will introduce other problems.
(Note: the particular settings of -10dB and 3:1 are just examples in this case)

Limiters, like Wave Hammer, usually apply high ratios from 10:1 through to infinity:1 (brick wall limiting) and therefore apply much greater gain reduction.

The track plug-ins in Vegas are normally defaulted to off and setting them on doesn't make them actually do anything, as their settings default to null positions, until you change setting on the plug-ins or load a preset (which changes the settings).

the only change I made in Hammer was to change the volume threshold to -12.6, and applied that to all audio events

Did you apply Wave Hammer as a non-realtime effect to each of the events?

If so you have actually recorded new audio events in place of the originals. While this is a possible way of working it means it is difficult to undo the changes made. It is better to apply the effects as real time, track based events, and automate the settings, so you can adjust them to specific events, if you don't want the effect applied to all events on the track.
solarblast1 wrote on 12/11/2008, 9:12 PM
I used Non-RT to each event; however, if I had known that I could do it on the track, that would have simplified matters. Actually, keeping track of Takes wasn't too bad, and it's pretty easy to see what of a number of Takes on an event do. Just press T, and one can cycle through them. I'll look at the automation of (audio) events later. There may be a place for them on my next project.

The compression ideas are beginning to sink in now between these messages, Vegas Help, and the Wikipedia entry. What I've done so far actually sounds pretty good. I did take note of the gain as 3:1 and boosting the volume, as in another post a few back. I haven't noticed any distortion in the audio track, and even some bothersome pops have become tolerable.

I haven't really found much depth on sound concepts and use in the Vegas manual or a Vegas book (Vegas Revealed, which is a good intro to Vegas otherwise) I have. I've been picking my way through this. It strikes me that one needs an audio engineer background to fathom some of these plug-ins. Maybe that's provided by some tutorial elsewhere. Possibly the AES has some sort of primer? Some time ago, a friend loaned me a book by Douglas Spotted Eagle, and it might have a chapter on AE basics. The same friend recently told me about reference signals that may be useful in connection with audio work. Possibly found in VAAST. Well, there's a lot to explore, but I think I'm about to produce the final rendering, and turn to some other needs. Another less demanding video effort looms down the road but some months away.