Hi, this is a very interesting project and I am wondering if you might have some insights into an experiment that it inspired me to persue

[mcanthony]

Would it be feasible to use this implementation / it's modelling techniques for recognition in "reverse" (ya I am a newb) as a method of better guiding a procedural generator of rain sounds? I suspect rain generators are a long researched subject and there may be a universally effective method of accomplishing a quality rain generator, perhaps you can point me in the right direction?

I have not got too deep into physical modelling but I suspect that is a likely direction, if so do you think parts of this software may actually add value (if abused correctly) to a physical modelling rain sound generator? I have the same curiosities about your wind detection algorithms, but rain seemed like the bigger fish.

Thanks for your time and nice work, with this and your BBC research, and with generally educating the worlds future!

[kenders2000]

The rain detector is built on real sounds so there isn’t a simulator per say. however for the wind noise algorithm I wrote a couple of simulators.

I haven't looked into rain simulation, but it is likely to be fairly straight forward, just superimposing lots of impulses on top of one another, with different filters for different impacts.

Hope that helps let me know if you want to look at the wind noise simulator, though it will take me time to extract and sort out.

Paul

--  Paul Kendrick Gmail Sent with Airmail

On 11 November 2015 at 08:38:14, Michael Anthony (notifications@github.com) wrote:

Thanks for your time and nice work, with this and your BBC research, and with generally educating the worlds future!