Imagine your cellphone rings. It's tiny speaker is vibrating in and out, producing sound waves that have this form:
Image Source : Tutorvista.com
Because the speaker is just repeating the same tone, unless you have it set to a song, the pattern of sound waves will be a longitudinal wave that has areas where the sound is strongest (compressions), and then areas where the sound is weakest (rarefactions).
It's pretty much the same effect when you answer the phone, but now you are making a more complicated set of compressions and rarefactions because you are talking with far more random sound effects than the speaker produces.
As the sound waves travel through the air, they heat it up very slightly, and that's where the energy that started in your lungs and pushed air through your vocal cords ends up, heating the air around you.
I certainly have heard a lot about the conservation of energy (of all forms) however when it comes to sound in particular, I just can't figure out how it could possibly be recycled into another energy form. When sound (rings of energy vibrating air molecules) is sent, how could that energy get turned into something else? Sure a tiny bit could hit something and turn into a tiny amount of kinetic energy, but doesn't the vast majority of it just fly off into oblivion?
Depending on how dense the air is, the vibrating air coming out of your mouth as you speak will be carried to other molecules, with the further molecules vibrating less as the energy is diluted as heat.
It does fly off into oblivion, which is lucky when its a voice you particularly don't want to hear, all sounds around you get mixed together from different sources and fade away.
So energy is conserved, but what was a concentrated burst of energy when you spoke ends ups as shaking up a large bulk of air molecules, all by a tiny, tiny amount.
My point is that energy is not lost, but rather spread out further and further until it can do no useful work.