@inproceedings{ma-etal-2026-s2m,
title = "{S}2{M}-Inject: Prompt-Guided Zero-Shot Voice Cloning for Music Generation",
author = "Ma, Teng and
Qiang, Chunyu and
Liu, Zhongliang and
Wang, Xiaopeng and
Yin, Kang and
Zhang, Ziyu and
Liang, Yuzhe and
Chen, Yushen and
Guo, Yuxin and
Chen, Feiyu",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.296/",
pages = "6282--6294",
ISBN = "979-8-89176-462-0",
abstract = "Despite advancements in controllable text-to-music (TTM) generation, existing models struggle with cross-domain voice cloning for singing due to inadequate modeling of speech-derived vocal characteristics, inherently limiting precise timbre transfer. To achieve precise and controllable music generation conditioned on the timbre of reference speech, we propose S2M-Inject, a cross-domain framework driven by reference speech. Incorporating a fixed timbre embedding strategy, S2M-Inject injects speaker representations into an instruction-guided generation pipeline. This effectively preserves target vocal attributes while enabling comprehensive natural-language control over lyrics, genre, instrumentation, rhythm, melody, and mood. Extensive experiments demonstrate that the framework achieves effective cross-domain voice cloning, significantly enhancing vocal similarity while maintaining high-fidelity audio quality in both Chinese and English tasks. This work extends voice cloning into the musical domain, offering a rigorous paradigm for personalized music generation. Audio samples are available at https://anonymous.4open.science/w/S2M-Inject-975."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ma-etal-2026-s2m">
<titleInfo>
<title>S2M-Inject: Prompt-Guided Zero-Shot Voice Cloning for Music Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Teng</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chunyu</namePart>
<namePart type="family">Qiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhongliang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaopeng</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kang</namePart>
<namePart type="family">Yin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ziyu</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuzhe</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yushen</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuxin</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Feiyu</namePart>
<namePart type="family">Chen</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Despite advancements in controllable text-to-music (TTM) generation, existing models struggle with cross-domain voice cloning for singing due to inadequate modeling of speech-derived vocal characteristics, inherently limiting precise timbre transfer. To achieve precise and controllable music generation conditioned on the timbre of reference speech, we propose S2M-Inject, a cross-domain framework driven by reference speech. Incorporating a fixed timbre embedding strategy, S2M-Inject injects speaker representations into an instruction-guided generation pipeline. This effectively preserves target vocal attributes while enabling comprehensive natural-language control over lyrics, genre, instrumentation, rhythm, melody, and mood. Extensive experiments demonstrate that the framework achieves effective cross-domain voice cloning, significantly enhancing vocal similarity while maintaining high-fidelity audio quality in both Chinese and English tasks. This work extends voice cloning into the musical domain, offering a rigorous paradigm for personalized music generation. Audio samples are available at https://anonymous.4open.science/w/S2M-Inject-975.</abstract>
<identifier type="citekey">ma-etal-2026-s2m</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.296/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>6282</start>
<end>6294</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T S2M-Inject: Prompt-Guided Zero-Shot Voice Cloning for Music Generation
%A Ma, Teng
%A Qiang, Chunyu
%A Liu, Zhongliang
%A Wang, Xiaopeng
%A Yin, Kang
%A Zhang, Ziyu
%A Liang, Yuzhe
%A Chen, Yushen
%A Guo, Yuxin
%A Chen, Feiyu
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F ma-etal-2026-s2m
%X Despite advancements in controllable text-to-music (TTM) generation, existing models struggle with cross-domain voice cloning for singing due to inadequate modeling of speech-derived vocal characteristics, inherently limiting precise timbre transfer. To achieve precise and controllable music generation conditioned on the timbre of reference speech, we propose S2M-Inject, a cross-domain framework driven by reference speech. Incorporating a fixed timbre embedding strategy, S2M-Inject injects speaker representations into an instruction-guided generation pipeline. This effectively preserves target vocal attributes while enabling comprehensive natural-language control over lyrics, genre, instrumentation, rhythm, melody, and mood. Extensive experiments demonstrate that the framework achieves effective cross-domain voice cloning, significantly enhancing vocal similarity while maintaining high-fidelity audio quality in both Chinese and English tasks. This work extends voice cloning into the musical domain, offering a rigorous paradigm for personalized music generation. Audio samples are available at https://anonymous.4open.science/w/S2M-Inject-975.
%U https://aclanthology.org/2026.findings-emnlp.296/
%P 6282-6294
Markdown (Informal)
[S2M-Inject: Prompt-Guided Zero-Shot Voice Cloning for Music Generation](https://aclanthology.org/2026.findings-emnlp.296/) (Ma et al., Findings 2026)
ACL
- Teng Ma, Chunyu Qiang, Zhongliang Liu, Xiaopeng Wang, Kang Yin, Ziyu Zhang, Yuzhe Liang, Yushen Chen, Yuxin Guo, and Feiyu Chen. 2026. S2M-Inject: Prompt-Guided Zero-Shot Voice Cloning for Music Generation. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 6282–6294, Budapest, Hungary. Association for Computational Linguistics.