Granular effects
Granular processing of audio in real-time is a tricky proposition. Unlike granular synthesis, there are no samples to prepare, loop or manipulate, so applying the concept of ‘grains’ to a live input will either introduce excessive latency or constrain the possibilities in comparison to sample-based synthesis tools.
These considerations aren’t necessarily limitations, though. Rather, they’re simply a different way of approaching signal processing in order to break audio into tiny chunks for further manipulation. For example, using LFObased amplitude modulation, you can easily slice the audio input into smaller pulses – as long as you’re using waveforms that ‘chop’ the sound, like squares, pulses and modified sawtooth shapes.
Another common processing trick for creating granular effects on-the-fly is to use delays, often in conjunction with the above amplitude modulation method. What’s more, by modulating various delay parameters, it’s possible to enhance the chaotic complexity that’s often associated with the sound of granular synthesis. It’s just a matter of deciding which elements you want to put into motion.
Finally, it may surprise some producers to learn that they might well have already been applying granular processing to their audio without realising it. That is, the earliest tools for time-stretching audio were actually based on grains! In the 80s and early 90s, these were originally offline processes that applied granular algorithms to sampled data. Nowadays, a quick review of Live’s warp modes will reveal parameters for Grain Size and Flux, which can even be automated within clips. We’ll explore those in the walkthrough below.