@inproceedings{li-etal-2026-tts,
title = "{TTS}-Hub: Leveraging Modular {L}o{RA}s and Arithmetic Composition for Controllable Text-to-Speech",
author = "Li, Xiang and
Zhang, Shiqi and
Xu, Zichen and
Gu, Wenyuan and
Xiao, Hongru and
Cheng, Bo and
Zhou, Li and
Han, Jiale and
Wang, Benyou",
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.202/",
pages = "4613--4639",
ISBN = "979-8-89176-459-0",
abstract = "Controllable text-to-speech (TTS) aims to generate speech from text with precise control over prosodic and speaker-related attributes such as pitch, age, and accent. Existing methods mainly rely on natural language prompts or reference audio cloning. However, prompt-based approaches often suffer from the cross-modal semantic gap between textual descriptions and target speech attributes, resulting in imprecise and coarse-grained control, while cloning-based methods are constrained by reference audio and offer limited flexibility. To address these limitations, we propose TTS-Hub, a controllable TTS framework that leverages modular Low-Rank Adaptation (LoRA) components and their arithmetic composition for fine-grained and flexible speech control. Specifically, we first build a comprehensive Data Hub covering 6 high-level categories and 32 fine-grained speech attributes, and use it to construct a LoRA Hub in which each LoRA is specialized for a specific attribute. During inference, TTS-Hub selects and composes the required LoRA modules through simple arithmetic operations, enabling flexible and extensible attribute control without retraining the backbone model. Experiments on two mainstream TTS frameworks demonstrate that TTS-Hub consistently outperforms strong baselines and enables precise and flexible control."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-tts">
<titleInfo>
<title>TTS-Hub: Leveraging Modular LoRAs and Arithmetic Composition for Controllable Text-to-Speech</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xiang</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shiqi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zichen</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenyuan</namePart>
<namePart type="family">Gu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hongru</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bo</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Li</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiale</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benyou</namePart>
<namePart type="family">Wang</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>Controllable text-to-speech (TTS) aims to generate speech from text with precise control over prosodic and speaker-related attributes such as pitch, age, and accent. Existing methods mainly rely on natural language prompts or reference audio cloning. However, prompt-based approaches often suffer from the cross-modal semantic gap between textual descriptions and target speech attributes, resulting in imprecise and coarse-grained control, while cloning-based methods are constrained by reference audio and offer limited flexibility. To address these limitations, we propose TTS-Hub, a controllable TTS framework that leverages modular Low-Rank Adaptation (LoRA) components and their arithmetic composition for fine-grained and flexible speech control. Specifically, we first build a comprehensive Data Hub covering 6 high-level categories and 32 fine-grained speech attributes, and use it to construct a LoRA Hub in which each LoRA is specialized for a specific attribute. During inference, TTS-Hub selects and composes the required LoRA modules through simple arithmetic operations, enabling flexible and extensible attribute control without retraining the backbone model. Experiments on two mainstream TTS frameworks demonstrate that TTS-Hub consistently outperforms strong baselines and enables precise and flexible control.</abstract>
<identifier type="citekey">li-etal-2026-tts</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.202/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>4613</start>
<end>4639</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TTS-Hub: Leveraging Modular LoRAs and Arithmetic Composition for Controllable Text-to-Speech
%A Li, Xiang
%A Zhang, Shiqi
%A Xu, Zichen
%A Gu, Wenyuan
%A Xiao, Hongru
%A Cheng, Bo
%A Zhou, Li
%A Han, Jiale
%A Wang, Benyou
%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 li-etal-2026-tts
%X Controllable text-to-speech (TTS) aims to generate speech from text with precise control over prosodic and speaker-related attributes such as pitch, age, and accent. Existing methods mainly rely on natural language prompts or reference audio cloning. However, prompt-based approaches often suffer from the cross-modal semantic gap between textual descriptions and target speech attributes, resulting in imprecise and coarse-grained control, while cloning-based methods are constrained by reference audio and offer limited flexibility. To address these limitations, we propose TTS-Hub, a controllable TTS framework that leverages modular Low-Rank Adaptation (LoRA) components and their arithmetic composition for fine-grained and flexible speech control. Specifically, we first build a comprehensive Data Hub covering 6 high-level categories and 32 fine-grained speech attributes, and use it to construct a LoRA Hub in which each LoRA is specialized for a specific attribute. During inference, TTS-Hub selects and composes the required LoRA modules through simple arithmetic operations, enabling flexible and extensible attribute control without retraining the backbone model. Experiments on two mainstream TTS frameworks demonstrate that TTS-Hub consistently outperforms strong baselines and enables precise and flexible control.
%U https://aclanthology.org/2026.emnlp-main.202/
%P 4613-4639
Markdown (Informal)
[TTS-Hub: Leveraging Modular LoRAs and Arithmetic Composition for Controllable Text-to-Speech](https://aclanthology.org/2026.emnlp-main.202/) (Li et al., EMNLP 2026)
ACL
- Xiang Li, Shiqi Zhang, Zichen Xu, Wenyuan Gu, Hongru Xiao, Bo Cheng, Li Zhou, Jiale Han, and Benyou Wang. 2026. TTS-Hub: Leveraging Modular LoRAs and Arithmetic Composition for Controllable Text-to-Speech. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 4613–4639, Budapest, Hungary. Association for Computational Linguistics.