Normalise (and other filters) a selection?

Noob here. How do I make a selection and then apply a filter to that selection? Parts of my wav file are quiet, for example, so I just want to high-light the quiet bits and normalise them. Currently I am having to split the wav file and normalise each new wav section by selecting Normalise Region, which is silly.

Any help gratefully received.

I think that’s the way you have to do it.

Another possibility, which in addition gives you more control over the amplification, is to use an effect that has a Makeup Gain (like the Tap Dynamics or the SC4) and automate that.
Keep it at zero during the loud parts and crank it up during the soft.

Along the same lines as Peder… you really have to just pick the best poison. If the stretch of low volume sound is rather long, sometimes splicing the track and normalizing it is the easist option. If the parts are all over the place (like getting control of double bass drums or even bass guitar), using the SC4 Compressor along the entire track is a must. Lastly, you can always use envelopes to make slight changes to dbs along the track path (or automate the gain).

another way is to use the envelope functions of a region. Show the envelope with “envelope visible” from the context menu and then change the volume (don’t forget “enable envelop”). I prefer this to automating the volume of a track.

Hmmm, why didn’t I know about envelopes before. Could’ve helped instead of having to tediously adjust the volume automation everytime I decided to raise the other levels…

That said, the pros of using Makeup Gain is that you have 20-24dBs to play with as opposed to the 6 the envelope has. But if that is enough for your track it’s by far the easiest.
And of course you could raise the track volume to +6 dB, use the envelope to attenuate the loud parts by the same amount and push the quiet parts by 6, thereby effectively increasing those by 12 dBs. That might possibly introduce some distortion in the louder parts though.

Sometimes I require up to 12db gain, so this makeup gain could be the way forward.

For the record I am dealing solely with spoken vocals (it’s for podcasting) so some vocals can be quieter than others. I’m often dealing with large wav files of up to 10 minutes so splicing just to make volume adjustments is a bit of a pain.

I do think all this sounds fiddly. In Audition as soon as you’ve highlighted a section a dial appears that allows instant gain adjustments. As you turn the knob you can see the wave file getting bigger or smaller instantly, and so therefore see visually whether you’re peaking or not. In fact this is a good way of manually normalising stuff, especially where there is a sudden bang or click that is too loud. Using this selection technique is very quick and is far less involved than envelopes.

@demonboy: its not fiddly at all once you enable the region gain envelope and make it visible. however, ardour deliberately does NOT redraw the waveform, because of a general committment to the idea that you should mix with your ears and not your eyes. normalization is even faster: select a region and press “n”. if you want to debate DAW design philosophy, there’s always time for a chat on IRC, but i find the forums a really tedious medium for such discussions.