On a larger setup, I guess multiple coils & magnets would be on the belts. The belt with magnets could be strung from the top of a tree and anchored to a spring on the ground, so it can "stretch" without snapping as the treetop sways. Coil's could be set on a seperate belt that parallels the first, with the spring at the opposite end, so the stretching of the second belt doesn't pull the coils in the same direction as the magnets.
Wouldn't adding a few blades to the belts cause more movement of the magnets and coils, leading to more production in the coils? I'm not talking about blades that spin smoothly. I'm talking about uneven blades made specifically to cause turbulence on the belt, shaking the heck out of the magnets and coils. Seems like the more turbulence, the more movement of the magnets and coils, the more electricity from the coils---in theory.
How much movement from how much wind seems to depend on what type of tree and how tall it is. Different types of trees sway differently in the wind.
Anyway, that's how I'd try to set up wind belts if I were attempting to power my home. If I had no trees, I guess I'd try a pole with a spring section near the top, to facilitate the sway.