Think I recall conversations with friends, or on dA or RAT, about making an infrasonic amp like that.
Not that anyone would have the power available to run it, or the transformers or speakers to make use of it, let alone the program matter to amplify with it. 
I suppose hydrogen thyratrons could do reasonable audio-range output, but an OPT is still required -- the pulsed current ratings might be high, but continuous ratings matter here, and the voltage is quite high (~kV) too. To get high efficiency and power output while still meeting ratings, you need about the same load impedance as any other tube (i.e. a kohm or thereabouts). So it's not really all that interesting.
Tim
EXCITRONS DAMMIT, I don't care one bit about thyratron amps, I dont care about class D amps. the whole reason I made this thread is because Excitrons (NOT IGNITRONS OR THYRATRONS) have a grid that can control current flow linearly, unlike thyratrons which the grid acts like an on-switch, and to reapeat, this thread is about Excitrons!
with the right excitron in theory you shouldn't even need an output transformer, just make it directly coupled to the speaker, and someone must've made an exitron that was small enough to be practical for an audio amp, one that didnt deliver kiloamps to the speaker if you accidentally touched the input jack inducing hum into it

again, EXCITRONS,
NOT THYRATRONS, Excitron excitron excitron!!!
thyratrons are like a GDT with the grid being able to trigger them.
while excitrons are like MVR's with a grid and exciter circuit, so you don't need to "trigger" them to get current flow, and the grid can be used to linearly control the current.
they're kinda like a gas filled triode with a constant excitation circuit flowing current trough the gas to keep it ionized.