Lune

EMNLP Findings2020

TurnGPT: a Transformer-based Language Model for Predicting Turn-taking in Spoken Dialog

Erik Ekstedt, Gabriel Skantze

Abstract

Syntactic and pragmatic completeness is known to be important for turn-taking prediction, but so far machine learning models of turn-taking have used such linguistic information in a limited way. In this paper, we introduce TurnGPT, a transformer-based language model for predicting turn-shifts in spoken dialog. The model has been trained and evaluated on a variety of written and spoken dialog datasets. We show that the model outperforms two baselines used in prior work. We also report on an ablation study, as well as attention and gradient analyses, which show that the model is able to utilize the dialog context and pragmatic completeness for turn-taking prediction. Finally, we explore the model's potential in not only detecting, but also projecting, turn-completions.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

7 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Investigating the Representation of Backchannels and Fillers in Fine-tuned Language Models cites From text to talk: Harnessing conversational corpora for humane and diversity-aware language technology
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers7

  1. Proactive Conversational Agents with Inner ThoughtsXingyu Bruce Liu, Shitao Fang, Weiyan Shi, Chien-Sheng Wu et al.CHI 2025 · 76 citations
  2. Predicting Turn-Taking and Backchannel in Human-Machine Conversations Using Linguistic, Acoustic, and Visual SignalsYuxin Lin, Yinglin Zheng, Ming Zeng, Wangzheng ShiACL 2025 · 5 citations
  3. Speculative End-Turn Detector for Efficient Speech Chatbot AssistantHyunjong Ok, Suho Yoo, Jaeho LeeACL 2026 · 5 citations
  4. Investigating the Representation of Backchannels and Fillers in Fine-tuned Language ModelsYu Wang, Leyi Lao, Langchu Huang, Gabriel Skantze et al.ACL 2026 · 1 citation
  5. From text to talk: Harnessing conversational corpora for humane and diversity-aware language technologyMark Dingemanse, Andreas LiesenfeldACL 2022

Related papers

  1. Response-conditioned Turn-taking PredictionBing'er Jiang, Erik Ekstedt, Gabriel SkantzeACL Findings 2023 · 3 citations
  2. Large Language Models Know What To Say But Not When To SpeakMuhammad Umair, Vasanth Sarathy, Jan Peter de RuiterEMNLP Findings 2024 · 6 citations
  3. Talking Turns: Benchmarking Audio Foundation Models on Turn-Taking DynamicsSiddhant Arora, Zhiyun Lu, Chung-Cheng Chiu, Ruoming Pang et al.ICLR 2025
  4. DialogBERT: Discourse-Aware Response Generation via Learning to Recover and Rank UtterancesXiaodong Gu, Kang Min Yoo, Jung-Woo HaAAAI 2021 · 83 citations
  5. AV-Dialog: Spoken Dialogue Models with Audio-Visual InputTuochao Chen, Bandhav Veluri, Hongyu Gong, Shyamnath GollakotaACL 2026 · 1 citation
Dusk over the sea between two cliffs drawn in fine vertical lines