Why does transcription confuse voices?

(Unbelievable...THIRD time I’ve been thru this process of editing my question and the retyping my context. Who designed this dialog, the Department of Redundancy Department?)

Anyway, I transcribed a sequence, and Transcription can’t distinguish between a thirty-something, deep voiced person and an older thin voiced male. Trust me, their voices are different.

Is there a way to “tune” the AI so it figures out that there are two speakers here? Right now, the transcription lists any speaker as “Speaker 2”