@inproceedings{jiang-etal-2026-speechrole,
title = "{S}peech{R}ole: A Large-Scale Dataset and Benchmark for Evaluating Speech Role-Playing Agents",
author = "Jiang, Changhao and
Sun, Jiajun and
Cao, Yifei and
Zhuang, Jiabao and
Che, Xinmeng and
Li, Hui and
Fan, Xiaoran and
Zhang, Ming and
Ye, Junjie and
Dou, Shihan and
Xi, Zhiheng and
Tong, Jingqi and
Wu, Yilong and
Fan, Baoyu and
Ji, Tao and
Gui, Tao and
Zhang, Qi and
Huang, Xuanjing",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.2196/",
pages = "49929--49954",
ISBN = "979-8-89176-459-0",
abstract = "Speech is essential for realistic role-playing, yet existing work on role-playing agents largely centers on text, leaving Speech Role-Playing Agents (SRPAs) underexplored and without systematic evaluation. We introduce SpeechRole, a unified framework for developing and assessing SRPAs. SpeechRole-Data contains 98 roles and 111k speech-to-speech conversations with rich timbre and prosodic variation, providing large-scale resources for training SRPAs. SpeechRole-Eval offers a multidimensional benchmark that directly evaluates generated speech, preserving paralinguistic cues and measuring interaction ability, speech expressiveness, and role-playing fidelity. Experiments show that end-to-end SRPAs such as GPT-4o Audio achieve strong fluency and naturalness, but remain limited in prosody consistency and emotion appropriateness. Current open-source end-to-end models exhibit clear performance gaps, particularly in interaction ability and role-playing fidelity, while several cascaded systems achieve relatively stable performance on these dimensions. Furthermore, role-specific training can further improve the role-playing capabilities of end-to-end SRPAs."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="jiang-etal-2026-speechrole">
<titleInfo>
<title>SpeechRole: A Large-Scale Dataset and Benchmark for Evaluating Speech Role-Playing Agents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Changhao</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yifei</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiabao</namePart>
<namePart type="family">Zhuang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinmeng</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hui</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoran</namePart>
<namePart type="family">Fan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ming</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junjie</namePart>
<namePart type="family">Ye</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shihan</namePart>
<namePart type="family">Dou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiheng</namePart>
<namePart type="family">Xi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingqi</namePart>
<namePart type="family">Tong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yilong</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Baoyu</namePart>
<namePart type="family">Fan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tao</namePart>
<namePart type="family">Ji</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tao</namePart>
<namePart type="family">Gui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanjing</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Speech is essential for realistic role-playing, yet existing work on role-playing agents largely centers on text, leaving Speech Role-Playing Agents (SRPAs) underexplored and without systematic evaluation. We introduce SpeechRole, a unified framework for developing and assessing SRPAs. SpeechRole-Data contains 98 roles and 111k speech-to-speech conversations with rich timbre and prosodic variation, providing large-scale resources for training SRPAs. SpeechRole-Eval offers a multidimensional benchmark that directly evaluates generated speech, preserving paralinguistic cues and measuring interaction ability, speech expressiveness, and role-playing fidelity. Experiments show that end-to-end SRPAs such as GPT-4o Audio achieve strong fluency and naturalness, but remain limited in prosody consistency and emotion appropriateness. Current open-source end-to-end models exhibit clear performance gaps, particularly in interaction ability and role-playing fidelity, while several cascaded systems achieve relatively stable performance on these dimensions. Furthermore, role-specific training can further improve the role-playing capabilities of end-to-end SRPAs.</abstract>
<identifier type="citekey">jiang-etal-2026-speechrole</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.2196/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>49929</start>
<end>49954</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SpeechRole: A Large-Scale Dataset and Benchmark for Evaluating Speech Role-Playing Agents
%A Jiang, Changhao
%A Sun, Jiajun
%A Cao, Yifei
%A Zhuang, Jiabao
%A Che, Xinmeng
%A Li, Hui
%A Fan, Xiaoran
%A Zhang, Ming
%A Ye, Junjie
%A Dou, Shihan
%A Xi, Zhiheng
%A Tong, Jingqi
%A Wu, Yilong
%A Fan, Baoyu
%A Ji, Tao
%A Gui, Tao
%A Zhang, Qi
%A Huang, Xuanjing
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F jiang-etal-2026-speechrole
%X Speech is essential for realistic role-playing, yet existing work on role-playing agents largely centers on text, leaving Speech Role-Playing Agents (SRPAs) underexplored and without systematic evaluation. We introduce SpeechRole, a unified framework for developing and assessing SRPAs. SpeechRole-Data contains 98 roles and 111k speech-to-speech conversations with rich timbre and prosodic variation, providing large-scale resources for training SRPAs. SpeechRole-Eval offers a multidimensional benchmark that directly evaluates generated speech, preserving paralinguistic cues and measuring interaction ability, speech expressiveness, and role-playing fidelity. Experiments show that end-to-end SRPAs such as GPT-4o Audio achieve strong fluency and naturalness, but remain limited in prosody consistency and emotion appropriateness. Current open-source end-to-end models exhibit clear performance gaps, particularly in interaction ability and role-playing fidelity, while several cascaded systems achieve relatively stable performance on these dimensions. Furthermore, role-specific training can further improve the role-playing capabilities of end-to-end SRPAs.
%U https://aclanthology.org/2026.emnlp-main.2196/
%P 49929-49954
Markdown (Informal)
[SpeechRole: A Large-Scale Dataset and Benchmark for Evaluating Speech Role-Playing Agents](https://aclanthology.org/2026.emnlp-main.2196/) (Jiang et al., EMNLP 2026)
ACL
- Changhao Jiang, Jiajun Sun, Yifei Cao, Jiabao Zhuang, Xinmeng Che, Hui Li, Xiaoran Fan, Ming Zhang, Junjie Ye, Shihan Dou, Zhiheng Xi, Jingqi Tong, Yilong Wu, Baoyu Fan, Tao Ji, Tao Gui, Qi Zhang, and Xuanjing Huang. 2026. SpeechRole: A Large-Scale Dataset and Benchmark for Evaluating Speech Role-Playing Agents. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 49929–49954, Budapest, Hungary. Association for Computational Linguistics.