On: Scientists say you can cancel the noise but keep y...

[ww520]

It's a hard problem. I wonder whether they can combine a specially made window that funnels the traveling of the sound wave to the active noise cancellation mic and speakers to make it easier.

A back of the envelope thought. A quad-pane window with the double-pane at outside to cut down the noise, a middle air gap, and then a double-pane inside. There's an opening at one end of the outside double-pane to allow air to go between the middle air gap and an opening at the other end of the double-pane inside to allow air to flow through. The ANC can be placed along the middle air gap to reduce the noise more.

       air in
    ===      ===================
    |   noise ->   <- ANC      | 
    ===================      ===
                       air out

It would be expensive but it's a first step.

[Anechoic → ww520]

First thing: unless that middle air gap is very wide (10+ inches), you've just created a resonant chamber that almost completely negates the sound reducing performance of the two double-pane layers. This is way party walls consisting of two separately framed walls (drywall-stud-drywall-gap-drywall-stud-drywall) tend to be very poor at blocking sounds between units.

There are other issues regarding the impedance mismatch between the outside and middle air gap openings.

As you said, it's a very hard problem.

[ww520 → Anechoic]

Can the resonant chamber actually help in the sense that it tunes the sound into certain frequency range? Making the job of the ANC easier? Also can the glass material be tuned to reduce the resonance? Music instruments need to be make certain way to take advantage of the resonant chamber effect; any deviation would produce no sound.