IIRC, the affected family members have impairments in both the cognitive and motor part of language, but I don't think the motor defecit is causing the cognitive defecit. You could get around orofacial motor impairment by using sign language. After some googling, I couldn't tell whether or not people have tried to teach sign language to the affected KE family members, but that seems an obvious solution. If it had worked, we would've heard about it by now, so probably there's more to the story than motor difficulties.

The paper about ultrasonic "songs" in mice made quite a splash, but like many things in science, it was not really that new.

It has been known for 30+ years that rats make ultrasonic vocalizations, that their variation & quality are correlated with reproductive fitness, and that their production varies depending on the hormonal state of potential mates. This is essentially what the mouse paper was all about, yet the basic phenomenon had been in print for quite a long time.

That's not to say mice (& rats) have something to tell us about language. I think they do, particularly the role of preoptic, hypothalamic and midbrain centers involved in the production of emotional utterances. These areas are involved in rodent vocalizations, and may be in humans too-- particularly for non-verbal output such as laughter, crying, moaning, etc.

As for Buzaki's remarks about motor output being important in sensory map development, I think he is absolutely right and I don't think it stops with the more well-known topographic maps seen in cortex and other laminar structures. I am willilng to bet that emotional expressions also serve to wire up the networks involved in the recognition of social cues. There may be maps that lay between say, the olfactory cortex and preoptic/hypothalamic/midbrain output nuclei, but they don't have a intelligible topology, so far as we presently know.