
Podcast Audiograms Must Not Invent A Speaker Reaction
A podcast audiogram can quietly add a claim the recording never made. A generated face smiles after a difficult quote. A nod appears before the guest answers. A reaction shot lands under words spoken by someone else. Viewers read those movements as part of the conversation even when they were created later. Treat audio timing, speaker identity and visible reaction as separate evidence. The recording establishes who said what and when. A transcript locates the quote. Generated motion may support the clip, but it cannot establish a facial response that the source camera never captured. Video…
The skinny
The skinny isn't ready yet — notes appear once the transcript is processed.