In the process of extending the transform I developed, it occurred to me that the sound that is absorbed by the skull is translated and could technically change the neural response based on the kinetic activity at the nerve junctions. Probably not a great effect, but nevertheless possible. I am testing a model of the head with vertices as points in a matrix transform. The origin momentum comes from the difference in pressure between lungs and the atmosphere. Each point in a surface has a different temporal response. It does seem that four models that can undergo a transform could define the entire process. What is new is the possibility of using ray tracing within the GPU to simulate the sound. If the listener is the camera and the light source is the vocal cords, or the differential pressure, then it is possible to have the GPU do all the work of generating a voice from the light emitted from surfaces over time. If colors are used as momentum or frequency profiles, then as the light is varied to conform to the intensity of differential momentum, the reflections would emulate the produced sound.
It would only be necessary to simulate the form once and use the vector product model matrix to create the continuous transform.
So the light that was collected at the camera would be the equivalent of the sound produced. I will have to investigate the use of a GPU to do secondary reflection as well as variable transparency. It does seem there are three distinct modes which would be modeled differently. The rapid change in pressure, white noise, and base frequency resonance ( with various size, cavity resonators ). Each mode would have a different transform and the white noise and sharp pressure changes seem fairly obvious without any temporal vector analysis. Perhaps I can change the transform to be done in one pass by overlay of surfaces and reflections so that an area within the camera view would be summed as the pressure change that is measured by the ear.
There is another odd thing that goes on with the Milky Way galaxy and it would be interesting to see what models are being used and determine if they have considered all the normal variations which would happen due to temporal variance, momentum profiles, natural oscillations, and many other effects that would influence the process to act in very specific modes. This is the problem with conclusions without the worked out methods. It may be sufficient for entertainment science that sells copies of Scientific American, but I find it to be junk because I can't look to see if they have missed a step. It is somewhat like swine before pearls. It presumes the entire audience lacks the intellect to correct mistakes.