Modeling turn-taking with distant viewing: investigating silence thresholds in human and AI-generated discourse

2026/07/20 by Taylor Arnold, Nicolas Ballier, Artem Saloev
#cs.CL

paper · pdf

Abstract

This study investigates silence gaps in two kinds of audiovisual material. We analysed thirty US situational comedies and fifty-one synthetic podcasts generated with Google NotebookLM. Gaps were compared across speaker gender, assigned from a fundamental-frequency threshold estimated in Praat, and across production settings.

Citations

• A simplest systematics for the organization of turn-taking for conversation

Related

• Talking Turns: Benchmarking Audio Foundation Models on Turn-Taking Dynamics
• Audibility and Intelligibility of Talk in Passing-By Interactions: A Methodological Puzzle
• Moravec's Paradox: Towards an Auditory Turing Test
• Predicting Turn-Taking and Backchannel in Human-Machine Conversations Using Linguistic, Acoustic, and Visual Signals
• Dramatic Conversation Disentanglement