Prompt-Guided Turn-Taking Prediction

TL;DR

This paper introduces a transformer-based turn-taking prediction model that can be controlled via textual prompts, enabling dynamic adaptation to conversational contexts and improving prediction accuracy.

Contribution

The study presents a novel prompt-guided turn-taking prediction model that incorporates textual prompts into transformer architectures, allowing explicit control over conversational timing.

Findings

Improved turn-taking prediction accuracy with prompt-guided control.

Effective adaptation of turn-taking timing based on textual instructions.

Synthetic prompt sentences generated using LLMs facilitated training.

Abstract

Turn-taking prediction models are essential components in spoken dialogue systems and conversational robots. Recent approaches leverage transformer-based architectures to predict speech activity continuously and in real-time. In this study, we propose a novel model that enables turn-taking prediction to be dynamically controlled via textual prompts. This approach allows intuitive and explicit control through instructions such as "faster" or "calmer" adapting dynamically to conversational partners and contexts. The proposed model builds upon a transformer-based voice activity projection (VAP) model, incorporating textual prompt embeddings into both channel-wise transformers and a cross-channel transformer. We evaluated the feasibility of our approach using over 950 hours of human-human spoken dialogue data. Since textual prompt data for the proposed approach was not available in existing datasets, we utilized a large language model (LLM) to generate synthetic prompt sentences. Experimental results demonstrated that the proposed model improved prediction accuracy and effectively varied turn-taking timing behaviors according to the textual prompts.