Turn-taking is a basic organizational feature of human conversation and remains difficult to model in natural, synchronous dialog systems. While existing research has explored multimodal approaches and large language models for turn-ending prediction, there is a lack of naturalistic conversational corpora specifically addressing turn-taking dynamics in Turkish. This study introduces a multimodal Turkish conversational dataset of unscripted dyadic interactions, comprising synchronized front-facing video, per-speaker audio channels that allow overlapping speech to be attributed to individual speakers, and time-aligned transcriptions. Turn-taking prediction is formulated as a binary classification problem, and a Genetic Algorithm (GA) is employed to optimize interpretable decision rules derived from visual, acoustic, and linguistic features. A hybrid AND-OR rule representation is adopted in the proposed framework to represent the alternative cue combinations that precede a turn transition.</p>\n","updatedAt":"2026-08-27T06:13:26.335Z","author":{"_id":"6630eb15e65d0b18d0cfdbcb","avatarUrl":"/avatars/d89e31d6f16e2b2be337b02d719c2cee.svg","fullname":"Ahmet Tuğrul Bayrak","name":"tugrulbayrak","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9092802405357361},"editors":["tugrulbayrak"],"editorAvatarUrls":["/avatars/d89e31d6f16e2b2be337b02d719c2cee.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.22071","authors":[{"_id":"6a8fd51e3bd48bb654ea692b","user":{"_id":"6630eb15e65d0b18d0cfdbcb","avatarUrl":"/avatars/d89e31d6f16e2b2be337b02d719c2cee.svg","isPro":false,"fullname":"Ahmet Tuğrul Bayrak","user":"tugrulbayrak","type":"user","name":"tugrulbayrak"},"name":"Ahmet Tuğrul Bayrak","status":"claimed_verified","statusLastChangedAt":"2026-08-27T08:45:04.991Z","hidden":false},{"_id":"6a8fd51e3bd48bb654ea692c","name":"Fatma Nur Korkmaz","hidden":false},{"_id":"6a8fd51e3bd48bb654ea692d","name":"Bekir Berker Türker","hidden":false},{"_id":"6a8fd51e3bd48bb654ea692e","name":"Mustafa Sertaç Türkel","hidden":false},{"_id":"6a8fd51e3bd48bb654ea692f","name":"Alper Kaplan","hidden":false}],"publishedAt":"2026-08-22T00:00:00.000Z","submittedOnDailyAt":"2026-08-27T00:00:00.000Z","title":"Real-TurnTurk: A Multimodal Turkish Corpus for Turn-Taking Prediction","submittedOnDailyBy":{"_id":"6630eb15e65d0b18d0cfdbcb","avatarUrl":"/avatars/d89e31d6f16e2b2be337b02d719c2cee.svg","isPro":false,"fullname":"Ahmet Tuğrul Bayrak","user":"tugrulbayrak","type":"user","name":"tugrulbayrak"},"summary":"Turn-taking is a basic organizational feature of human conversation and remains difficult to model in natural, synchronous dialog systems. While existing research has explored multimodal approaches and large language models for turn-ending prediction, there is a lack of naturalistic conversational corpora specifically addressing turn-taking dynamics in Turkish. This study introduces a multimodal Turkish conversational dataset of unscripted dyadic interactions, comprising synchronized front-facing video, per-speaker audio channels that allow overlapping speech to be attributed to individual speakers, and time-aligned transcriptions. Turn-taking prediction is formulated as a binary classification problem, and a Genetic Algorithm (GA) is employed to optimize interpretable decision rules derived from visual, acoustic, and linguistic features. A hybrid AND-OR rule representation is adopted in the proposed framework to represent the alternative cue combinations that precede a turn transition.","upvotes":0,"discussionId":"6a8fd51f3bd48bb654ea6930","ai_summary":"A multimodal Turkish dialogue dataset and genetic algorithm-optimized interpretable rules are used to predict turn transitions from visual, acoustic, and linguistic cues.","ai_keywords":["turn-taking prediction","multimodal dataset","Genetic Algorithm","AND-OR rule representation","visual features","acoustic features","linguistic features","Turkish conversational corpus"],"ai_summary_model":"thinkingmachines/Inkling-Small"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.22071.md","query":{}}">
Real-TurnTurk: A Multimodal Turkish Corpus for Turn-Taking Prediction
Abstract
A multimodal Turkish dialogue dataset and genetic algorithm-optimized interpretable rules are used to predict turn transitions from visual, acoustic, and linguistic cues.
Turn-taking is a basic organizational feature of human conversation and remains difficult to model in natural, synchronous dialog systems. While existing research has explored multimodal approaches and large language models for turn-ending prediction, there is a lack of naturalistic conversational corpora specifically addressing turn-taking dynamics in Turkish. This study introduces a multimodal Turkish conversational dataset of unscripted dyadic interactions, comprising synchronized front-facing video, per-speaker audio channels that allow overlapping speech to be attributed to individual speakers, and time-aligned transcriptions. Turn-taking prediction is formulated as a binary classification problem, and a Genetic Algorithm (GA) is employed to optimize interpretable decision rules derived from visual, acoustic, and linguistic features. A hybrid AND-OR rule representation is adopted in the proposed framework to represent the alternative cue combinations that precede a turn transition.
Community
Turn-taking is a basic organizational feature of human conversation and remains difficult to model in natural, synchronous dialog systems. While existing research has explored multimodal approaches and large language models for turn-ending prediction, there is a lack of naturalistic conversational corpora specifically addressing turn-taking dynamics in Turkish. This study introduces a multimodal Turkish conversational dataset of unscripted dyadic interactions, comprising synchronized front-facing video, per-speaker audio channels that allow overlapping speech to be attributed to individual speakers, and time-aligned transcriptions. Turn-taking prediction is formulated as a binary classification problem, and a Genetic Algorithm (GA) is employed to optimize interpretable decision rules derived from visual, acoustic, and linguistic features. A hybrid AND-OR rule representation is adopted in the proposed framework to represent the alternative cue combinations that precede a turn transition.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2608.22071 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.22071 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.