@inproceedings{zou-etal-2026-lts,
title = "{LTS}-{V}oice{A}gent: A Listen-Think-Speak Framework for Efficient Streaming Voice Interaction via Semantic Triggering and Incremental Reasoning",
author = "Zou, Wenhao and
Miao, Yuwei and
Ma, Zhanyu and
Xu, Jun and
Gao, Jiuchong and
He, Renqing and
Xu, Jingwen",
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.1712/",
pages = "39049--39069",
ISBN = "979-8-89176-459-0",
abstract = "Real-time voice agents face a tension: end-to-end models often lack deep reasoning, while cascaded ASR{--}LLM{--}TTS pipelines achieve stronger reasoning but defer LLM inference until after ASR finalizes the transcript, creating an avoidable post{--}turn-end silent gap. We target this gap under complex spoken requests that admit a relatively determinate answer, and treat it as system-induced rather than communicatively motivated. Existing cascaded streaming strategies attempt to reduce it via mechanical segmentation (e.g., fixed chunks, VAD-based splitting) or speculative generation, but they frequently either break semantic units or waste computation on predictions that must be rolled back. We propose LTS-VoiceAgent, a Listen{--}Think{--}Speak framework that explicitly separates when to think from how to reason incrementally. It features a Dynamic Semantic Trigger to detect meaningful prefixes, and a Dual-Role Stream Orchestrator that coordinates a background Thinker (for state maintenance) and a foreground Speaker (for speculative solving). This parallel design enables ``thinking while listening'' without committing audible content before turn end. We also introduce a Pause-and-Repair benchmark that injects natural disfluencies as a semantic-level streaming stress test. Experiments across VERA, Spoken-MQA, BigBenchAudio, and our benchmark show that LTS-VoiceAgent achieves a stronger accuracy{--}latency{--}efficiency trade-off than serial cascaded baselines and existing streaming strategies."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zou-etal-2026-lts">
<titleInfo>
<title>LTS-VoiceAgent: A Listen-Think-Speak Framework for Efficient Streaming Voice Interaction via Semantic Triggering and Incremental Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wenhao</namePart>
<namePart type="family">Zou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuwei</namePart>
<namePart type="family">Miao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhanyu</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiuchong</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Renqing</namePart>
<namePart type="family">He</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingwen</namePart>
<namePart type="family">Xu</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>Real-time voice agents face a tension: end-to-end models often lack deep reasoning, while cascaded ASR–LLM–TTS pipelines achieve stronger reasoning but defer LLM inference until after ASR finalizes the transcript, creating an avoidable post–turn-end silent gap. We target this gap under complex spoken requests that admit a relatively determinate answer, and treat it as system-induced rather than communicatively motivated. Existing cascaded streaming strategies attempt to reduce it via mechanical segmentation (e.g., fixed chunks, VAD-based splitting) or speculative generation, but they frequently either break semantic units or waste computation on predictions that must be rolled back. We propose LTS-VoiceAgent, a Listen–Think–Speak framework that explicitly separates when to think from how to reason incrementally. It features a Dynamic Semantic Trigger to detect meaningful prefixes, and a Dual-Role Stream Orchestrator that coordinates a background Thinker (for state maintenance) and a foreground Speaker (for speculative solving). This parallel design enables “thinking while listening” without committing audible content before turn end. We also introduce a Pause-and-Repair benchmark that injects natural disfluencies as a semantic-level streaming stress test. Experiments across VERA, Spoken-MQA, BigBenchAudio, and our benchmark show that LTS-VoiceAgent achieves a stronger accuracy–latency–efficiency trade-off than serial cascaded baselines and existing streaming strategies.</abstract>
<identifier type="citekey">zou-etal-2026-lts</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.1712/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>39049</start>
<end>39069</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LTS-VoiceAgent: A Listen-Think-Speak Framework for Efficient Streaming Voice Interaction via Semantic Triggering and Incremental Reasoning
%A Zou, Wenhao
%A Miao, Yuwei
%A Ma, Zhanyu
%A Xu, Jun
%A Gao, Jiuchong
%A He, Renqing
%A Xu, Jingwen
%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 zou-etal-2026-lts
%X Real-time voice agents face a tension: end-to-end models often lack deep reasoning, while cascaded ASR–LLM–TTS pipelines achieve stronger reasoning but defer LLM inference until after ASR finalizes the transcript, creating an avoidable post–turn-end silent gap. We target this gap under complex spoken requests that admit a relatively determinate answer, and treat it as system-induced rather than communicatively motivated. Existing cascaded streaming strategies attempt to reduce it via mechanical segmentation (e.g., fixed chunks, VAD-based splitting) or speculative generation, but they frequently either break semantic units or waste computation on predictions that must be rolled back. We propose LTS-VoiceAgent, a Listen–Think–Speak framework that explicitly separates when to think from how to reason incrementally. It features a Dynamic Semantic Trigger to detect meaningful prefixes, and a Dual-Role Stream Orchestrator that coordinates a background Thinker (for state maintenance) and a foreground Speaker (for speculative solving). This parallel design enables “thinking while listening” without committing audible content before turn end. We also introduce a Pause-and-Repair benchmark that injects natural disfluencies as a semantic-level streaming stress test. Experiments across VERA, Spoken-MQA, BigBenchAudio, and our benchmark show that LTS-VoiceAgent achieves a stronger accuracy–latency–efficiency trade-off than serial cascaded baselines and existing streaming strategies.
%U https://aclanthology.org/2026.emnlp-main.1712/
%P 39049-39069
Markdown (Informal)
[LTS-VoiceAgent: A Listen-Think-Speak Framework for Efficient Streaming Voice Interaction via Semantic Triggering and Incremental Reasoning](https://aclanthology.org/2026.emnlp-main.1712/) (Zou et al., EMNLP 2026)
ACL