The most important thing in the notice is that they actually did and experiment:

The Duke researchers randomly extracted over 100,000 speech samples, each 0.1 second long, from recordings of thousands of English sentences. Acoustic analysis of the combined samples revealed 10 frequency peaks that match the most significant intervals used in musical scales worldwide.

The result confirms (if test conditions and data reduction methods were appropriate) what has been generally assumed, and has been found in other tests.

It is generally also held (with some experimental confirmation) that the same result would be obtained from a similar test of bird calls, animal grunts, and virtually any other "natural" complex source.

The "interval matching" observed by the Duke researcers has, in fact, been applied to studies of "communication" between animals. In very simplistic terms, if two animals "harmonize" it is (by some researchers) assumed that they are communicating with each other, if there is no "common interval" relationship, then it's more likely just that they felt like making a noise.

The research performed, and the result obtained, appear to be a worthwhile bit of work. The anthropomorphization evidenced in the "announcement," by trying to "prove" that humans are "the source of everything," went out of style several hundred years ago. What is really "proved," perhaps, is that the human vocal tract has evolved to produce, with some facility, the intervals that are "easiest" to produce and that are easiest to distinguish (and hence best for carrying information content?).

Of course, the tone of their announcement is well suited to attract attention, perhaps, from a broader range of persons. It's a forgivable bit of "adspeak," if their full publication shows accurately what they did and what result they obtained.

John