Awesome. Such progress.
The paper’s at:
but can we do it in blender?
you do not need complex maths to do that , any modern VA synth with FM synthesis can do similar sounds.
Just google ALESIS ANDROMEDA analog synthesizer pouring water sound.
Synthesizing it and making it physically accurate are two very different things.
I have actually been wondering if there was such a thing as computer-rendered sound. Imagine what it would sound like hitting Suzanne or something. The tricky part about such things is that natural substances are often heterogeneous, so sounds like hitting wood or wind blowing through trees would be very difficult to synthesize. I guess that’s why they did water, at least it is a homogeneous.
If you liked that you have, no you must, hear the job of xoxos
listen to his human , animals , machines synthetisers , awesome , he is a ganius for me.
actually it is not , and they can never be . I guess what you are talking about is about “Physical modeling” which by the way is a form of sound synthesis . Physical modeling uses a specific mathematical model to imitate a specific sound and thus is much more accurate.
Computer rendered sounds imitating real ones have been done in CSOUND for decades now and they are nothing new.
nice link tweakingknobs , will check it out tomorrow thoroughly.
CSOUND can interface with blender perfectly cause as Blender has a python api .
That paper and videos aren’t about how to synthesize a sound. They describe a process how to render the sound, caused by interaction of simulated bodies - in this case pouring water. You do not know what specific sound you need, as this specific sound depends on dinamically changing bodies. So that paper is a more about determining what sound will correspond the situation, rather than how to mix/render sounds.