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

Taylor Arnold; Nicolas Ballier; Artem Saloev

cs.CLJul 20, 2026

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

Authors: Taylor Arnold, Nicolas Ballier, Artem Saloev

Organizations: Data Science and Statistics, University of Richmond, Virginia, U.S.A. · ALTAE, Université Paris Cité, Paris, France

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.

Explore similar work

ParaPairAudioBench: Paralinguistic Pairwise Audio Benchmark for LALM-as-a-Judge
Jun 23, 2026Jisu Jeon, Seungyeon Jwa, Joosung Lee +6Paralinguistic Cues Large Audio Language Models
Low-Latency Real-Time Audio Game Commentary System via LLM-Based Parallel Text Generation
Jun 11, 2026Ryota Kawamatsu, Anum Afzal, Yuki Saito +5Joint Audio-Video Generation Audio
When Entanglement Lower-Bounds Disparity: Auditing and Repairing Demographic Fairness in Audio Understanding Models
Sep 23, 2026Kian Shamsaie, Iman Modarressi

Jun 23, 2026 · cs.SDJ/K move · Enter open · S save

ParaPairAudioBench: Paralinguistic Pairwise Audio Benchmark for LALM-as-a-Judge

Jisu Jeon, Seungyeon Jwa, Joosung Lee, Jinhyeon Kim+5

Hongik University, South Korea · Seoul National University, South Korea · NAVER Cloud, South Korea · KAIST, South Korea

Large Audio-Language Models (LALMs) have been widely used as judge models for the automatic evaluation of generated speech. However, prior approaches predominantly focus on holistic naturalness, leaving fine-grained paralinguistic distinctions underexplored. We introduce ParaPairAudioBench, a pairwise benchmark of 5,175 audio pairs across five paralinguistic dimensions: Style, Rate, Emphasis, Age, and Gender. Our experiments show that current LALM judges still lag behind human judgments by 32%p on average and exhibit severe calibration failures, particularly in Tie cases where the correct decision is to abstain. To further analyze lexical versus acoustic reliance, the benchmark includes both same-transcript and cross-transcript conditions. ParaPairAudioBench enables multi-dimensional, calibration-aware assessment of the reliability of LALM-as-a-Judge for paralinguistic speech evaluation.

Paralinguistic Cues Large Audio Language Models Llm-As-A-Judge Multicalibration and Calibration Spatial Transcriptomics

Open paper Ask about it PDF

Abstract

Explore similar work

ParaPairAudioBench: Paralinguistic Pairwise Audio Benchmark for LALM-as-a-Judge

Low-Latency Real-Time Audio Game Commentary System via LLM-Based Parallel Text Generation

When Entanglement Lower-Bounds Disparity: Auditing and Repairing Demographic Fairness in Audio Understanding Models