AgentSteerTTS: A Multi-Agent Closed-Loop Framework for Composite-Instruction Text-to-Speech

Bin Kang 1,2,3, Shaoguo Wen 3, Yang Fan 2, Shunlong Wu 4, Junjie Wang 2, Yulin Li 2, Junzhi Zhao 5, Junle Wang 3, Zhuotao Tian 2,† · 2026

Read AgentSteerTTS: A Multi-Agent Closed-Loop Framework for Composite-Instruction Text-to-Speech by Bin Kang 1,2,3, Shaoguo Wen 3, Yang Fan 2, Shunlong Wu 4, Junjie Wang 2, Yulin Li 2, Junzhi Zhao 5, Junle Wang 3, Zhuotao Tian 2,† (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-17583.txt.

Description

Bin Kang and colleagues present AgentSteerTTS, a multi-agent closed-loop frame for intent-faithful expressive steering of composite speech instructions. A disentanglement agent separates identity from emotion-prosody spaces to stop speaker-emotion leakage. A dual-stream anchor control grounds intents with retrieved acoustic anchors fused into control vectors. A fast-slow feedback agent refines intensity by latent gradient fixes. It also resolves meaning-sound mismatches by perceptual review. Tests show steady significant gains over baselines.

Catalog record

CREATORBin Kang 1,2,3, Shaoguo Wen 3, Yang Fan 2, Shunlong Wu 4, Junjie Wang 2, Yulin Li 2, Junzhi Zhao 5, Junle Wang 3, Zhuotao Tian 2,†
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE48300
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.17583

Browse the full library · About the Tower of Babel