Because sounds are made by shapes. The shapes of our mouths and the shapes of objects colliding. And sounds have shapes, shapes over time.

Maybe because when you pronounce a B, it's very hard to make it sound sharp. The lips rolling off one another naturally have a gentle ramp-up. It's really hard to have a sudden transient. You can record this in a free DAW and see it in front of you.

By contrast, it's very hard to make a K- sound without a sharp transient.

With B you are adding pressure to two soft tissues in contact.

With K you are adding pressure when one tissue is hard, and saliva is present (causing a relatively sudden production of the sound).

Try making B sound sharp, hard to do while saying Bouba.

Try making K sound soft, basically impossible, especially while saying Kiki.

Either similar underlying processing machinery (due to auditory and visual processing evolving from simpler and more interconnected processing circuits) or because when thinking of frequencies of waveforms in light/sound, visual sharpness and auditory sharpness really are the same thing (so even if they didn't co-evolve, it can be more efficient or useful to have some single underlying representation waveform representation that represents these things similarly).

tbf for me the Bouba evokes curvature because it makes me think of a Baobab tree. Not so sure about the dogs :)