However inefficient (or bizarre) this may sound— can a line-level 60Hz sine (originating from a sound card, oscillator, etc) amplified by a heavy duty 1000W+ RMS monoblock be used as a substitute for a high voltage AC power source?
If the square root of (1000W * 4 ohms) gets us 63V, AND supposing we use a step up transformer on that signal to bring us into the 120V range, theoretically, would we end up with a usable 126V 7.9A (from amps = watts / volts) power source?
Yes.
For a class A-B amplifier it wouldn't be a very efficient as much power would be lost in the output transistors. You could also run into power-factor problems if driving an inductive load. The out-of-phase voltage and current could damage the output transistors at high power.
In addition to TRANSISTOR's answer. Actually you will obtain still an amplifier. Any kind of noise, caused by nearby phone, by lightning, by soundcard's DAC, by bad wire contact, will be reflected as voltage burst (amplified, in other words), which most likely will break your consumer or an amplifier, or will cause a lack of consumer's working stability. So only a consumer, not sensitive to dirty voltage and to overload, is allowed to use - like some kind of lamps, a electro-magnets, a sources of heat. If you still want to test your devices combination, I would recommend you at least to add 60Hz bandpass filter after soundcard/oscillator. But generally this is a bad idea, except of satisfaction of your intellectual curiosity.
P.S. Regarding to dirty voltage. Please read about measurement of sound amplifiers audio power. Your 1000W RMS is measured in according to standard procedure - 1kHz of input signal. At this frequency the 4Ohm-speaker actually will have near 200-300 Ohm of complex resistance. The permanent working with only 4-8Ohm of active resistance may overheat your amplifier.
