Like Ken mention in their answer, the speed of sound does NOT determine the amplitude of the sound. This means that you can't really rely on one to make any conclusions about the other (at least not in a direct way).
Additionally (like Ken also mentioned), in the first case (room boundaries present) the volume will be lower in the position of B because some of the energy (the amount depends on many factors such as the material(s) and composition of the boundaries, the frequency of sound, the way the boundaries are connected - flanking transmission - and some more) is reflected back into the room where A is positioned and some is dissipated as heat inside the material of the walls.
Reflection takes place due to the impedance mismatch of the media (air and wall material) and dissipation, mostly due to internal loses at the media (most of the loss mechanisms are a bit complex to discuss here).
Now, Stratiev gave a nice hint on clarifying a bit the given answers by providing an indirect question on the connection between the medium in which sound travels (this also determines the speed of sound) and the volume.
Assuming homogeneous media (this may not be a valid assumption for water when the travelling distance is quite long and under certain conditions), the speed of sound is constant as well as the loss mechanisms. Again, what determines the volume difference at the position of B in the two cases (assuming the same radiated power with the same directivity of course) are the loses exhibited in each medium. Thus, once more, the speed of sound is not directly related to the volume at the position of B for a homogeneous medium and for constant environmental conditions.