The distinction you (the OP) appear to be drawing between "positive current" being sent to the speaker via one conductor, and electrons being returned to the amplifier via the other conductor, is not correct.
During half-cycles of the AC waveform in which the voltage applied by the amplifier on its + output terminal is positive with respect to its - output terminal electrons in the + conductor of the cable move **very** slowly in the direction toward the amp’s + terminal, while an equal number of other electrons in the - conductor of the cable move **very** slowly toward the speaker. (Many electrons in both conductors also move in random directions at all times, but the movements I am referring to that occur in response to the voltage applied by the amplifier can be thought of as a **net** movement of electrons).
During the next half-cycle, when the voltage applied by the amplifier on its + output terminal is negative with respect to its - output terminal electrons in the two conductors move **very** slowly in the opposite of those directions, on a net basis.
Therefore individual electrons don’t go from one end of the cable to the other. They just move back and forth over a very small distance in response to the AC voltage applied by the amp.
Meanwhile, energy and power corresponding to the signal is conveyed in the form of an electromagnetic wave which moves from the amplifier to the speakers at near light speed, usually somewhere between 50% and 98% or so of the speed of light in a vacuum.
The very fast propagation of that electromagnetic wave and the very slow back and forth movement of electrons in both conductors are intimately related, and neither would exist without the other in the case of signals being conducted via wires.
Most of the energy and power being conveyed by the electromagnetic wave propagates through the dielectric material which surrounds and separates the two conductors, i.e., the insulation. That is why the specific velocity at which a signal propagates in a given cable type, within that range of 50% to 98% or so of the speed of light in a vacuum, depends on the "dielectric constant" aka "relative permittivity" of the particular insulation. A very small fraction of the electromagnetic wave propagates into the conductors, though, corresponding to the fraction of its energy and power that is absorbed by the resistance of the conductors. Power being energy per unit time.
Hope that clarifies more than it confuses :-)
Regards,
-- Al