@inproceedings{yang-etal-2026-nushuvoice,
title = {{N}{\"u}shu{V}oice: Reviving the Voice of Endangered N{\"u}shu with Pitch-Aware Text-to-Speech},
author = "Yang, Hongkun and
Xinhui, YI and
Xiyan, Zhao and
Meng, Yibo and
Wang, Lionel Z. and
Wang, Lixu and
Zhang, Yaqi and
Chen, Ruiqi and
Zhao, Xuanyue and
Zhang, Lanxin and
Zeng, Yu and
Weijia, Chu and
Ma, Yiming and
Liu, Chenyu and
Lin, Jianghao and
Xin, Xu",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {IJCNLP}-{AACL} 2026",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-aacl.123/",
pages = "2423--2434",
ISBN = "979-8-89176-444-6",
abstract = {N{\"u}shu is an endangered phonetic script historically used by women in Jiangyong County, southern Hunan, China. While existing computational studies of N{\"u}shu mainly focus on textual digitization and visual recognition, the acoustic reconstruction of its authentic pronunciation remains largely unexplored. Building a N{\"u}shu text-to-speech (TTS) system is particularly challenging because available recordings are extremely limited and mostly consist of isolated syllable-level pronunciations rather than natural sentence-level utterances. In this work, we introduce N{\"u}shuVoice, the first TTS benchmark for N{\"u}shu. We construct a sentence-level N{\"u}shu text-to-audio dataset that aligns standardized Unicode N{\"u}shu text, phonetic transcriptions, standard Chinese translations, and archival recordings. To synthesize speech under this extreme low-resource setting, we propose N{\"u}shu-PitchVITS, an -conditioned VITS framework that leverages N{\"u}shu{'}s five-level pitch notation as an explicit prosodic inductive bias. Experimental results show that N{\"u}shu-PitchVITS outperforms strong TTS baselines in spectral fidelity, pitch reconstruction, and human-rated intelligibility. We publicly release the dataset and code at: https://anonymous.4open.science/r/Nvshu-TTS-2EB6.}
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-etal-2026-nushuvoice">
<titleInfo>
<title>NüshuVoice: Reviving the Voice of Endangered Nüshu with Pitch-Aware Text-to-Speech</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hongkun</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Y</namePart>
<namePart type="given">I</namePart>
<namePart type="family">Xinhui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhao</namePart>
<namePart type="family">Xiyan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yibo</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lionel</namePart>
<namePart type="given">Z</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lixu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaqi</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruiqi</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanyue</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lanxin</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Zeng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chu</namePart>
<namePart type="family">Weijia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yiming</namePart>
<namePart type="family">Ma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chenyu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianghao</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xu</namePart>
<namePart type="family">Xin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-444-6</identifier>
</relatedItem>
<abstract>Nüshu is an endangered phonetic script historically used by women in Jiangyong County, southern Hunan, China. While existing computational studies of Nüshu mainly focus on textual digitization and visual recognition, the acoustic reconstruction of its authentic pronunciation remains largely unexplored. Building a Nüshu text-to-speech (TTS) system is particularly challenging because available recordings are extremely limited and mostly consist of isolated syllable-level pronunciations rather than natural sentence-level utterances. In this work, we introduce NüshuVoice, the first TTS benchmark for Nüshu. We construct a sentence-level Nüshu text-to-audio dataset that aligns standardized Unicode Nüshu text, phonetic transcriptions, standard Chinese translations, and archival recordings. To synthesize speech under this extreme low-resource setting, we propose Nüshu-PitchVITS, an -conditioned VITS framework that leverages Nüshu’s five-level pitch notation as an explicit prosodic inductive bias. Experimental results show that Nüshu-PitchVITS outperforms strong TTS baselines in spectral fidelity, pitch reconstruction, and human-rated intelligibility. We publicly release the dataset and code at: https://anonymous.4open.science/r/Nvshu-TTS-2EB6.</abstract>
<identifier type="citekey">yang-etal-2026-nushuvoice</identifier>
<location>
<url>https://aclanthology.org/2026.findings-aacl.123/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>2423</start>
<end>2434</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T NüshuVoice: Reviving the Voice of Endangered Nüshu with Pitch-Aware Text-to-Speech
%A Yang, Hongkun
%A Xinhui, Y. I.
%A Xiyan, Zhao
%A Meng, Yibo
%A Wang, Lionel Z.
%A Wang, Lixu
%A Zhang, Yaqi
%A Chen, Ruiqi
%A Zhao, Xuanyue
%A Zhang, Lanxin
%A Zeng, Yu
%A Weijia, Chu
%A Ma, Yiming
%A Liu, Chenyu
%A Lin, Jianghao
%A Xin, Xu
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-444-6
%F yang-etal-2026-nushuvoice
%X Nüshu is an endangered phonetic script historically used by women in Jiangyong County, southern Hunan, China. While existing computational studies of Nüshu mainly focus on textual digitization and visual recognition, the acoustic reconstruction of its authentic pronunciation remains largely unexplored. Building a Nüshu text-to-speech (TTS) system is particularly challenging because available recordings are extremely limited and mostly consist of isolated syllable-level pronunciations rather than natural sentence-level utterances. In this work, we introduce NüshuVoice, the first TTS benchmark for Nüshu. We construct a sentence-level Nüshu text-to-audio dataset that aligns standardized Unicode Nüshu text, phonetic transcriptions, standard Chinese translations, and archival recordings. To synthesize speech under this extreme low-resource setting, we propose Nüshu-PitchVITS, an -conditioned VITS framework that leverages Nüshu’s five-level pitch notation as an explicit prosodic inductive bias. Experimental results show that Nüshu-PitchVITS outperforms strong TTS baselines in spectral fidelity, pitch reconstruction, and human-rated intelligibility. We publicly release the dataset and code at: https://anonymous.4open.science/r/Nvshu-TTS-2EB6.
%U https://aclanthology.org/2026.findings-aacl.123/
%P 2423-2434
Markdown (Informal)
[NüshuVoice: Reviving the Voice of Endangered Nüshu with Pitch-Aware Text-to-Speech](https://aclanthology.org/2026.findings-aacl.123/) (Yang et al., Findings 2026)
ACL
- Hongkun Yang, YI Xinhui, Zhao Xiyan, Yibo Meng, Lionel Z. Wang, Lixu Wang, Yaqi Zhang, Ruiqi Chen, Xuanyue Zhao, Lanxin Zhang, Yu Zeng, Chu Weijia, Yiming Ma, Chenyu Liu, Jianghao Lin, and Xu Xin. 2026. NüshuVoice: Reviving the Voice of Endangered Nüshu with Pitch-Aware Text-to-Speech. In Findings of the Association for Computational Linguistics: IJCNLP-AACL 2026, pages 2423–2434, Hengqin, China. Association for Computational Linguistics.