It speaks a language that any intelligent race could conceive. I would not consider math to be invented by humans. It was more like discovery through our interpretation using symbols and signs.
That reminds me of Max Tegmark's mathematical universe hypothesis, which posits that the universe is a mathematical structure, and that mathematicians are akin to archaeologists uncovering hidden or forgotten truths. Perhaps it's just semantics to attempt to distinguish this from some form of "true" creativity, but I like the idea that even our most obscure mathematical theories can't be completely decoupled from physical reality.
He goes further to state the Computable Universe Hypothesis, which "posits that all computable mathematical structures (in Gödel's sense) exist". Consequently, perhaps the existence of absurdly large, computable numbers (see this
for some examples) that are seemingly too large to describe anything in our own universe hints at the existence of a multiverse where they may find some use. (That said, some disturbingly large numbers seem to emerge from physical considerations in combinatorics from time to time...)
As a counterargument to the idea that math is adept at describing the world, there are quite a few elegant structures that "should" be exact, but are actually approximations. For instance,
is correct to at least 42 billion decimal places (but is provably not equal to