A sound is an accumulation of frequencies. For example, if you play a single note (= single frequency), your audio signal will look like this, a regular sine wave:
In a more complicated sound, many frequencies contribute to the final sound over time, and you end up with something like this:
An audio sample is one point of this representation of the sound. In other words, it is the sound amplitude at a given point in time. Then a sound is nothing more than the set of audio samples that compose it, from beginning to end.
Is volume affecting audiosamples?
Yes. The absolute value of one sample is the volume of the sound at this point in time. For example, if you look at the sound above, you can see that it begins very quietly, and then it suddenly becomes loud, etc.
Can I manipulate audio samples in real time without recharing the soundbuffer with a new manipulated file?
It depends on what you want to do exactly.
I don't know if these explanations make things clearer for you, I did my best but I admit that it is not simple to understand. I also hope that I didn't say incorrect things