Don’t worry too much about that. Many of those will be just notes or harmonics thereof. If turn up the peak hold to infinite or set it to average over a longer time you’ll see that tend to smooth out as long as there’s some harmonic movement in the piece. Maybe watch out for places where peaks stand up way above everything else or the troughs are really deep. Be aware that this kind of analysis tends to have less resolution in the lower frequencies, so those bumps come more frequently and more narrow as things get higher jusr as an artifact of how it’s done.
In the end, it doesn’t matter what it looks like. All that anybody cares about is how it sounds. If it sounds good, it will probably conform generally to something like pink noise, and if it’s way off it might indicate something you’re not hearing accurately, but completely do not let these visuals drive too many decisions directly.
can you please let me know about the peaks and valleys in this final waveform
is it bad to have many peaks next to each other very closely?
is it bad to have many wide valleys?
what else should i look out for?
i'm just trying to understand better what i see in terms of peaks and valleys...what can you see there?
Thanks
well, its all about frequency masking: peaks mask frequency content left and right to it!
Maybe you increase the smoothing resolution of your FFT to better identify peaks and valleys. And using a 3 or 4.5dB FFT slope makes reading easier.
Next step would be to level out peaks and valleys, based on the musical content in the specific frequency range. This can done with an EQ or any other frequency matching tool like Euphonia.
EQ graph....tonal balance
Linear Mode
