Search papers, labs, and topics across Lattice.
The paper introduces Syn-TurnTurk, a novel synthetic dataset for turn-taking prediction in Turkish dialogues, addressing the lack of resources for this task. The dataset was generated using Qwen LLMs to simulate realistic conversations with overlaps and strategic silences. Experiments using BI-LSTM and Ensemble (LR+RF) models demonstrate the dataset's effectiveness, achieving 0.839 accuracy and 0.910 AUC scores in predicting turn-taking.
Turkish chatbots can now handle turn-taking with human-like fluency, thanks to a new synthetic dataset that teaches them to predict conversational cues.
Managing natural dialogue timing is a significant challenge for voice-based chatbots. Most current systems usually rely on simple silence detection, which often fails because human speech patterns involve irregular pauses. This causes bots to interrupt users, breaking the conversational flow. This problem is even more severe for languages like Turkish, which lack high-quality datasets for turn-taking prediction. This paper introduces Syn-TurnTurk, a synthetic Turkish dialogue dataset generated using various Qwen Large Language Models (LLMs) to mirror real-life verbal exchanges, including overlaps and strategic silences. We evaluated the dataset using several traditional and deep learning architectures. The results show that advanced models, particularly BI-LSTM and Ensemble (LR+RF) methods, achieve high accuracy (0.839) and AUC scores (0.910). These findings demonstrate that our synthetic dataset can have a positive affect for models understand linguistic cues, allowing for more natural human-machine interaction in Turkish.
