FabFilter User Forum
Feature request: fundamental/overtone following and harmonic/noisy separation
I kinda had a similar idea for the first one :)
For a while now I isolate the fundamental deepest note of the vocal but for adding exciting from that root note to create cleaner harmonics only related to this one note. So an automatic detection would come in really handy for sure!!!
This has all been done before in SoundRadix' Surfer EQ (and other copycat fundamental-tracking EQ) and Eventide's SplitEQ (and other noise/tone separating FX), and others.... if you wanted to check those out.
Ah, I think I've heard of Surfer EQ before. That seems to be otherwise just what I want except that it's not a dynamic EQ so you can't easily compress the fundamental/overtone. I guess it might be somewhat possible with some parallel processing where you clone a track, set one to cut and other to boost and then compress the latter.
SplitEQ (and the new ozone) have transient/sustain mode. However, I think it is slightly different from noise/harmonic. One splits based on dynamics, the other by frequency analysis. But I think Surfer EQ could again be used to try to achieve the same. Also Transmutator plugin has noise/harmonic split as one of the modes (haven't yet tried it, but just bought it, mainly for the purpose of being able to fade between two tracks such as snare top and bottom with one knob instead of two sliders).
Here's two somewhat related feature ideas that are relatively easy to implement but as far as I know uncommon or nonexistent among competing products.
Feature 1:
Being able to assign a band in Pro Q3 to follow either the fundamental frequency or an Nth harmonic overtone of a monotonic signal such as vocal or bass. I imagine being able to use e.g. dynamic EQ to compress the fundamental or a certain overtone instead of a static frequency without manually automating the frequency of the band could be really useful at times.
Feature 2:
Similar to how you can separately process mids/sides and in some products transients/sustain, how about being able to select harmonics/noise? On a complex polyphonic signals this could be difficult, but at least on a monophonic source it should be somewhat straightforward to split the detected fundamental and its overtones onto one channel and everything else onto another one. The user could then apply processing (be it EQ, compression or saturation) separately to e.g. bass notes vs string noise or vocal vowels vs consonants.