Generative AI Speaking-Time Regulation: The First Integrated Framework for Unidirectional and Bidirectional Individualized Speaking-Time Control in Human–AI Conversations

The increasing integration of generative artificial intelligence (AI) into conversational environments has created new requirements for managing the temporal structure of human–AI interactions. Although existing research has investigated conversational turn-taking, speech-duration control, and time allocation among human participants, the systematic regulation of speaking time independently assigned to human users and generative AI systems remains an insufficiently unified research direction. This paper introduces Generative AI Speaking-Time Regulation (GASTR), a configurable framework that establishes individualized temporal constraints for each participant in a generative AI conversation. The framework supports unidirectional regulation, in which speaking-time limits apply exclusively to either the human user or the generative AI system, and bidirectional regulation, in which both participants receive independently configurable time limits. Unlike conventional conversation-duration limits, GASTR distinguishes individual speaking duration, per-turn allocation, cumulative speaking budgets, and session-level temporal constraints. It further introduces configurable enforcement, countdown mechanisms, pre-limit notifications, automatic interruption, time-extension requests, pause-and-resume operations, and adaptive temporal allocation. The framework is designed to accommodate text-based conversations, spoken interactions, multimodal communication, and multi-agent conversational environments. To the best of our knowledge, this paper presents the first explicitly integrated conceptual framework centered on independently configurable speaking-time regulation for both human and generative AI participants, including unidirectional and bidirectional control within a unified architecture. This contribution is distinguished from existing work on controlling the duration of AI-generated speech, conventional turn-taking, and time allocation in human-only meetings. The paper establishes the conceptual foundations, operational definitions, architectural components, mathematical formulation, implementation principles, evaluation methodology, and ethical considerations necessary for developing speaking-time-regulated generative AI systems.

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-17
DOI
https://doi.org/10.5281/zenodo.24297640
Primary Topic
Speech and dialogue systems
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Generative AI Speaking-Time Regulation: The First Integrated Framework for Unidirectional and Bidirectional Individualized Speaking-Time Control in Human–AI Conversations

Jincheng Zhang
17 links
Zenodo (CERN European Organization for Nuclear Research)
Speech and dialogue systems
preprint

Generative AI Speaking-Time Regulation: The First Integrated Framework for Unidirectional and Bidirectional Individualized Speaking-Time Control in Human–AI Conversations

Jincheng Zhang
preprint en

Abstract

The increasing integration of generative artificial intelligence (AI) into conversational environments has created new requirements for managing the temporal structure of human–AI interactions. Although existing research has investigated conversational turn-taking, speech-duration control, and time allocation among human participants, the systematic regulation of speaking time independently assigned to human users and generative AI systems remains an insufficiently unified research direction. This paper introduces Generative AI Speaking-Time Regulation (GASTR), a configurable framework that establishes individualized temporal constraints for each participant in a generative AI conversation. The framework supports unidirectional regulation, in which speaking-time limits apply exclusively to either the human user or the generative AI system, and bidirectional regulation, in which both participants receive independently configurable time limits. Unlike conventional conversation-duration limits, GASTR distinguishes individual speaking duration, per-turn allocation, cumulative speaking budgets, and session-level temporal constraints. It further introduces configurable enforcement, countdown mechanisms, pre-limit notifications, automatic interruption, time-extension requests, pause-and-resume operations, and adaptive temporal allocation. The framework is designed to accommodate text-based conversations, spoken interactions, multimodal communication, and multi-agent conversational environments. To the best of our knowledge, this paper presents the first explicitly integrated conceptual framework centered on independently configurable speaking-time regulation for both human and generative AI participants, including unidirectional and bidirectional control within a unified architecture. This contribution is distinguished from existing work on controlling the duration of AI-generated speech, conventional turn-taking, and time allocation in human-only meetings. The paper establishes the conceptual foundations, operational definitions, architectural components, mathematical formulation, implementation principles, evaluation methodology, and ethical considerations necessary for developing speaking-time-regulated generative AI systems.

Zenodo (CERN European Organization for Nuclear Research)
Speech and dialogue systems
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon