How does delay-based pitch-shift work?
Hey! I guess this is a quite easy question since the technique exists pretty long: I want to build a pitch-shifter based on delay. How does it work? I assume folloeing:
- building a tape-delay (which changes pitch when delay-time is changed)
- sawtooth modulation of delay-time
Anything useful about my guess? Please help! Thank you!
yep you’re on the right track– it’s also related to the doppler effect, when an approaching sound-source’s apparent pitch increases (it’s delay time decreases because sound takes less time to reach the listener), and vice-versa
as the delayed repeats ‘move further away’ (ie, delay time increases) the period of the frequencies increases, creating the illusion of a fall in pitch and vice versa. It’s obvious how this works in the real world, but perceptually? I haven’t a clue! There’s a great example of how to achieve this effect in MaxMSP on youtube, with helpful explanations of relevant formulae etc, by ‘dude837′
There’s a good explanation in Puckette’s book, which you can view free at:
There’s a PD example, which is fairly easy to translate into Max here:
and the explanation here:
You can ignore the maths, and still get the gist of it!
Thank you, everyone!
Just one fact which actually made me think about all that:
In an interview MOUSE ON MARS say "These are two delays which run into feedback quickly. And with the speed of the feedback-rate you can control the pitch." – This really confused me. :)
on mars physics are different.