You would compress a spoken track as a rule, spikes or no spikes.

try this freedbie: http://loudmax.blogspot.com/
Drop the THRESHOLD if you want to bring things up to even them out, lower the OUTPUT as your max output. Like -3 or -6. Never distorts or clips.

But yes, you should be able to add points to the AUTOMATION line on the audio track to correct here and there.

If you're only after peak limiting, the RMS normalizer in Sound Forge is nifty for this.

Wavehammer in Vegas is a great limiter / make-it-louder plugin. It should be able to squash the peaks without adversely affecting the sound quality.

I usually manually lower the peaks in Wavelab, then normalize the whole thing, but that's a lot more work, and doing that kind of thing in Vegas would be very slow and tedious.

~Jay

Another vote for Wave Hammer. I usually apply it to all audio, setting the threshold to be just about even with the loudest peaks i want to preserve. Anything below this threshold is unaltered, while anything above it is reduced.

Thanks,

Good suggestions, simple process, drop the effect on the timeline, tweak the settings, save a preset, and then use it over and over for similar videos. That is my intent.

Musicvid,
I have Sound Forge Audio Studio, not even sure where I got it, maybe it came with Vegas, never used it, but it sits on my computer. It did not recognize my M2T files from my camera. Curious, can you open Sound Forge from within Vegas?

In Vegas:

Under Options / Preferences / Audio - Preferred Audio Editor, browse to the executable for Sound Forge Audio Studio, Strike OK.

Right Click on the audio on the timeline and select "Open a copy in Sound Forge..."

Never used Studio but it should work.

I believe SF Studio has RMS Norm as an option, and should open within Vegas as John Dennis describes. When you return to Vegas, the fixed event will show as a new take.

Also, the suggestions to use Wavehammer are not exactly chopped liver. It is powerful and versatile in the right hands.

John, Musicvid,

Thanks, got it working. I want to simplify this process as much as possible. I've got over 60 videos, audio is different, but entirely similar. These options, once I figure them out, should make this quick and easy and acceptable for my needs.

Wavehammer might be the way to go, those high end spikes I have in the audio are annoying, seemed to quiet them down a bit.

One more question, so there is a school of thought that compression for talking is generally used, rather than clipping, because it increases the lows and decreases the highs? The idea seems to make sense.

RMS Normalisation changes the gain of an event so it has a set RMS value for the entire event. Depending on how you configure the tool it will apply a limiter to peaks that would otherwise be clipped.

My first tool to deal with this kind of problem is Eq. As the person get closer to the microphone the amount of low frequency energy increases due to the proximity effect inherent in cardiod microphones. That alone makes them sound louder. Loud sounds will always sound loud. Applying a good helping of low frequency roll off can help. Of course the ideal solution when recording a talking head is microphone placement so they cannot get too close to the microphone.

The compressor in the track header is also good enough to help here as well. The 1.5:1 @ -24dB Preset is a good starting point. It is important to monitor the "Reduction" meter in the compressor while playing the track to ensure the Threshold is set so the compressor is doing its job. The Track FXs come after Event Gain but before Track Volume and any Volume Envelope, it helps no end to get your head around this.

Bob.

Barry, not sure what you mean by "clipping," because as you've no doubt heard a clipped waveform (top lopped off) sounds really harsh.

As for compression, I think it's common for speech because the audio level can vary significantly as the speaker gets slightly closer or further away from the mic.

For music, I have mixed feelings about compression. Using some compression makes listening in an auto much more pleasant (because the interior background noise level is so high), but for critical listening at home I usually don't use any.

I use a moderate amount of compression of 2:1 starting at -15db.

Pop tunes use a tremendous amount of compression; classical stuff, hardly any.

At least there are my observations. Others may have more experience and can chime in.