Help Compressing Vocal!

[msmeritt]

I've recorded a vocal into DP, and overall it's fine, except the few more bass notes in the singers range go noticeably softer than the rest. It seems to me there's got to be an easy way to apply compression to bring those up without having too much impact on the rest of the vocal. I can pinpoint the frequency in Hz of the point below which I need to boost the signal, so shouldn't I able to take advantage of, e.g., the Masterworks Compressor plugin, and add that frequency as a threshold, and set some settings, and voila? I just have *no* experience with compression and I'm at a loss about what to do. Any help would be appreciated!

[conleycd]

In a circumstance like this I think you should forget a compressor in a traditional way. I would just automate some fader rides over those quiet parts. If you have Melodyne they have a neat amplitude increase on the soundbite blob.

That is how I would approach it.

CC

[Tritonemusic]

In a circumstance like this I think you should forget a compressor in a traditional way. I would just automate some fader rides over those quiet parts.

+1

[Dwetmaster]

Even without the Fader automation. You can simply make your changes to the bite's volume directly...

[resolectric]

Use Multiband Compression.

[msmeritt]

Multiband compression is just what I'd thought, because I figured I could pinpoint that break where, below a particular note/frequency, the singer becomes soft, and I could set below that to boost, and above that to, well, not do so much. But I've just no idea really how to do it.

I think, though, now's the time for me to explore the graphic editing of volume automation. I've definitely used automation before but inserting events one at a time. This quickly became too clunky a way to do what I needed with this particular song, which is why I started considering compression. I hadn't tried graphic editing before, but as with everything, I'm learning new functionality in DP as the need arises, and the need appears to have arisen now, and it seems so far to be a really good way to do it. Not quite as fast as a compression setting might have been, but reasonably efficient.

Thanks to all.

[conleycd]

I don't really think multiband compression is the right tool for this 0- but I could be wrong. You are presumably having difficulty with certain pitches the vocalist is singing their lower range. I would as I mentioned use either fader automation or has someone else has said use soundbite volume (although for some reason I don't like soundbite volume - not sure why - I think it is because I don't trust that it is actually working).

You could get more complex and copy out those parts to a separate track and run that separate track louder. You could then bounce them together to one or leave them separate. Make sure you add some fades on the end of the soundbites if you are going to do that to avoid non-zero crossing cuts.

I still think give the fader or soundbite gain a try.

CC

[resolectric]

Multiband compression is just what I'd thought, because I figured I could pinpoint that break where, below a particular note/frequency, the singer becomes soft, and I could set below that to boost, and above that to, well, not do so much. But I've just no idea really how to do it
....

If you have a software MB Comp, such as Waves C4, try this:
> activate one single compression band (bypass the other three).
> start playing the vocal track "soloing" that frequency band.
> select the upper and lower frequency limits for that band. This is done by ear. When you are listening to just the portion of the vocal that you wish to expand, you have it right.
> adjust Threshold so that the compression touches the sounds that you want to bring up in volume.
> adjust attack and release to appropriate values for a vocal. I rarely, if ever, use fast Attack settings when compressing a vocal, and especially when the region you want to affect is in the low-frequency domain. Since you want to avoid "pumping" the vocal's bass content, the Release value should be adjusted with care too. The value depends on the quickness of the syllables but i would go for something between 130 and 350 ms. You really have to listen carefully.
> remove Solo from the compression band. You are now listening to the full vocal's spectrum with compression on the selected band.
> adjust the compression Ratio to a value that seems to bring out an even and natural content of the frequencies you want to bring up.
> adjust Gain in the selected compression Band by bringing it up (Gain is the Volume output for this compression band and not the final Gain of the Compressor, wich would affect all the frequencies).
This final step is just like equalizing, but, with a dynamic equalization.

Of course most of these adjustments are done by ear and all the values are subjective and program dependent.
There can be no Presets for this but i hope this helps.

[msmeritt]

Thanks to everybody for the helpful input. I decided to try fader automation using the graphic editor in Sequence view, and this went pretty well, much better than my previous attempts at fader automation inserting events or even using a controller. I also added some light compression just to smooth out the edges and boost the signal a bit. It came out pretty well overall. All the other bits of advice here, I'm really glad to know them and will surely take advantage of them at some point depending on the situation. Thanks again!