@inproceedings{lee-etal-2025-behavior,
title = "Behavior-{SD}: Behaviorally Aware Spoken Dialogue Generation with Large Language Models",
author = "Lee, Sehun and
Kim, Kang-wook and
Kim, Gunhee",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.484/",
pages = "9574--9593",
ISBN = "979-8-89176-189-6",
abstract = "Spoken dialogue involves behaviors like turn-taking, interruptions, filler words, and backchannels, which make interactions more natural and engaging but are often overlooked in language models. These models struggle to explicitly model these behavioral traits, resulting in a less natural and personalized communication style that aligns with user needs. To address this challenge, we make two key contributions. First, we introduce Behavior-SD, a large-scale dataset containing over 100K spoken dialogues (2,164 hours) annotated with various conversational behaviors, synthesized via LLMs to model diverse full-duplex interactions. Second, we propose BeDLM, the first dialogue model capable of generating natural conversations conditioned on specific behavioral and narrative contexts, supporting simultaneous contributions from both speakers. Through human evaluations and behavior-adherence metrics, we demonstrate that BeDLM outperforms baseline models in generating natural, coherent, and behaviorally rich dialogues. Our work opens new possibilities for developing behaviorally-aware dialogue systems that more closely mimic human conversational dynamics, enhancing user engagement and communication effectiveness."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lee-etal-2025-behavior">
<titleInfo>
<title>Behavior-SD: Behaviorally Aware Spoken Dialogue Generation with Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sehun</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kang-wook</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gunhee</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>Spoken dialogue involves behaviors like turn-taking, interruptions, filler words, and backchannels, which make interactions more natural and engaging but are often overlooked in language models. These models struggle to explicitly model these behavioral traits, resulting in a less natural and personalized communication style that aligns with user needs. To address this challenge, we make two key contributions. First, we introduce Behavior-SD, a large-scale dataset containing over 100K spoken dialogues (2,164 hours) annotated with various conversational behaviors, synthesized via LLMs to model diverse full-duplex interactions. Second, we propose BeDLM, the first dialogue model capable of generating natural conversations conditioned on specific behavioral and narrative contexts, supporting simultaneous contributions from both speakers. Through human evaluations and behavior-adherence metrics, we demonstrate that BeDLM outperforms baseline models in generating natural, coherent, and behaviorally rich dialogues. Our work opens new possibilities for developing behaviorally-aware dialogue systems that more closely mimic human conversational dynamics, enhancing user engagement and communication effectiveness.</abstract>
<identifier type="citekey">lee-etal-2025-behavior</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.484/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>9574</start>
<end>9593</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Behavior-SD: Behaviorally Aware Spoken Dialogue Generation with Large Language Models
%A Lee, Sehun
%A Kim, Kang-wook
%A Kim, Gunhee
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F lee-etal-2025-behavior
%X Spoken dialogue involves behaviors like turn-taking, interruptions, filler words, and backchannels, which make interactions more natural and engaging but are often overlooked in language models. These models struggle to explicitly model these behavioral traits, resulting in a less natural and personalized communication style that aligns with user needs. To address this challenge, we make two key contributions. First, we introduce Behavior-SD, a large-scale dataset containing over 100K spoken dialogues (2,164 hours) annotated with various conversational behaviors, synthesized via LLMs to model diverse full-duplex interactions. Second, we propose BeDLM, the first dialogue model capable of generating natural conversations conditioned on specific behavioral and narrative contexts, supporting simultaneous contributions from both speakers. Through human evaluations and behavior-adherence metrics, we demonstrate that BeDLM outperforms baseline models in generating natural, coherent, and behaviorally rich dialogues. Our work opens new possibilities for developing behaviorally-aware dialogue systems that more closely mimic human conversational dynamics, enhancing user engagement and communication effectiveness.
%U https://aclanthology.org/2025.naacl-long.484/
%P 9574-9593
Markdown (Informal)
[Behavior-SD: Behaviorally Aware Spoken Dialogue Generation with Large Language Models](https://aclanthology.org/2025.naacl-long.484/) (Lee et al., NAACL 2025)
ACL