EUVIMEDEUVIMED entdecken
Crossref

WenetSpeech-Yue: A Large-Scale Cantonese Speech Corpus with Multi-dimensional Annotation

Longhao Li, Zhao Guo, Hongjie Chen, Yuhang Dai, Ziyu Zhang, Hongfei Xue, Tianlun Zuo, Chengyou Wang, Shuiyuan Wang, Xin Xu, Hui Bu, Jie Li und 6 weitere

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (37) · 31627-31635

DOI: 10.1609/aaai.v40i37.40429 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Longhao Li und 17 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The development of speech understanding and generation has been significantly accelerated by the availability of large-scale, high-quality speech datasets. Among these, ASR and TTS are regarded as the most established and fundamental tasks. However, for Cantonese (Yue Chinese), spoken by approximately 84.9 million native speakers worldwide, limited annotated resources have hindered progress and resulted in suboptimal ASR and TTS performance. To address this challenge, we propose WenetSpeech-Pipe, an integrated pipeline for building large-scale speech corpus with multi-dimensional annotation tailored for speech understanding and generation. Based on this pipeline, we release WenetSpeech-Yue, the first large-scale Cantonese speech corpus with multi-dimensional annotation for ASR and TTS, covering 21,800 hours across 10 domains with annotations including ASR transcription, text confidence, speaker identity, age, gender, speech quality scores, among other annotations. We also release WSYue-eval, a comprehensive Cantonese benchmark with two components: WSYue-ASR-eval, a manually annotated set for evaluating ASR on short and long utterances, code-switching, and diverse acoustic conditions, and WSYue-TTS-eval, with base and coverage subsets for standard and generalization testing. Experimental results show that models trained on WenetSpeech-Yue achieve competitive results against state-of-the-art (SOTA) Cantonese ASR and TTS systems, including commercial and LLM-based models, highlighting the value of our dataset and pipeline.

Zitieren

APA

Li, L., Guo, Z., Chen, H., Dai, Y., Zhang, Z., Xue, H., Zuo, T., Wang, C., Wang, S., Xu, X., Bu, H., Li, J., Kang, J., Zhang, B., Yuan, R., Zhou, Z., Xue, W., & Xie, L. (2026). WenetSpeech-Yue: A Large-Scale Cantonese Speech Corpus with Multi-dimensional Annotation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(37), 31627-31635. https://doi.org/10.1609/aaai.v40i37.40429

Vancouver

Li L, Guo Z, Chen H, Dai Y, Zhang Z, Xue H, et al. WenetSpeech-Yue: A Large-Scale Cantonese Speech Corpus with Multi-dimensional Annotation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(37):31627-31635. doi:10.1609/aaai.v40i37.40429

BibTeX
@article{Li2026,
  author = {Li, Longhao and Guo, Zhao and Chen, Hongjie and Dai, Yuhang and Zhang, Ziyu and Xue, Hongfei and Zuo, Tianlun and Wang, Chengyou and Wang, Shuiyuan and Xu, Xin and Bu, Hui and Li, Jie and Kang, Jian and Zhang, Binbin and Yuan, Ruibin and Zhou, Ziya and Xue, Wei and Xie, Lei},
  title = {WenetSpeech-Yue: A Large-Scale Cantonese Speech Corpus with Multi-dimensional Annotation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {37},
  pages = {31627--31635},
  doi = {10.1609/aaai.v40i37.40429}
}
Kostenlos starten →