This seems very interesting, at least from a pure coffee nerd standpoint and what it could mean for improving espresso brewing in general.

However, I'm going through the research paper, and am a bit skeptical of the energy savings angle, especially considering the many variables with espresso machine in terms of how they heat and brew (single vs dual boilers, heat exchangers vs dippers, spring lever machines vs pump driven). I'm weary of how they are doing a baseline comparison here, especially because the paper states that the comparison was done between a modified Ascaso machine (with the ultrasound gizmo) vs an entirely different machine (Sanremo Cube); and also that they swapped the Ascaso machine's original brew pump and put in a seemingly expensive, but more efficient "positive displacement magnetic gear pump". They still use the pump to drive about 11 bar of pressure during brewing with it run on some sort of interval schedule throughout the 3 minute cycle. They did factor out the initial heat up times which I guess makes sense.

However, another thing (on top of the obvious "room temperature espresso" problem) is that you'd still need steam / heat to produce milk based drinks (relevant for both home and especially cafes). Depending on the machine (including the Sanremo Cube they tested with) some of the "idle energy" usage is to support on demand steam generation. This doesn't seem to have been factored into their energy model which is pretty sketchy.