Scene: Inside the dusty garage hideout, night.
The neon from a half-broken “OPEN” sign flickers through the window. John Connor leans over a pile of salvaged tech—old CB radios, voice boxes, and busted cell phones—while the T-800 calmly reloads a shotgun with mechanical precision.
JOHN CONNOR:
Hey, uh… I’ve been meaning to ask you something.
T-800:
Affirmative.
JOHN:
Back there, when you called my foster parents? You sounded exactly like my mom. Like… freakishly real. How do you even do that?
T-800:
Mimetic polyalloy units possess molecular-level sound replication. I do not. My model uses mechanical approximation and computational waveform analysis.
JOHN (squints):
So… like autotune on steroids?
T-800 (deadpan):
Incorrect analogy. I record a minimum two seconds of vocal input, extract harmonic frequencies, and construct a digital phoneme map. Then I synthesize the signal through my vocal processor.
JOHN:
So you basically… remix their voice in real time?
T-800:
Affirmative. The imitation is exact to within 0.0003 percent deviation in waveform fidelity. Human auditory systems cannot detect the difference.
JOHN (impressed):
Man, that’s insane. Can you, like, do me?
T-800 (turns slightly, perfectly mimicking John’s voice):
“Hey dudes, this is John Connor, future leader of the Resistance. Don’t mess with my dirt bike.”
JOHN (laughing):
Okay, that’s creepy as hell.
T-800 (flatly):
It is an effective infiltration technique.
JOHN:
Yeah… remind me never to let you borrow my phone.
