I think I would try with parallel interface ram just to ensure best performance. I would like to see if I could double the sample rate (obviously cuts down the max time of the delay).
Are there any specific chips you would recommend I use or is there a shield that exists that could be modified for this purpose?
What about this QuadRAM shield from Rugged http://ruggedcircuits.com/html/quadram.html
Also once this system is working I will try using a dedicated audio codec using the I2S interface.