Elamur Research⌕
☰
home / papers / 2026 / MauBERT: Universal Phonetic Inductive Biases for Few-Shot Acoustic Units Discovery

MauBERT: Universal Phonetic Inductive Biases for Few-Shot Acoustic Units Discovery

Paper · 2026 · Angelo Ortiz Tandazo, Manel Khentout, Youssef Benchekroun et al.

Summary

Self-supervised speech models such as HuBERT need very large amounts of clean audio. The units they discover are also shorter and more numerous than real phonemes, and they stay sensitive to speaker and phonetic context. This limits their use for low-resource and unwritten languages.

The authors propose MauBERT. They take an English-trained HuBERT-base and continue training it with phonetic supervision across 55 languages from the VoxCommunis Corpus, using articulatory feature annotations derived with PanPhon. There are two variants:

  • FEAT passes through an articulatory-feature bottleneck.
  • PHONE predicts phones directly.

To adapt to a new language, the model clusters its own representations and then fine-tunes itself on about 10 hours of unlabelled audio.

Evaluation uses ABX discriminability on ZRC2017, extended with five typologically varied languages, including Ukrainian. It also measures context invariance and compares read with casual speech. MauBERT produces more context-invariant representations than strong multilingual SSL baselines, and it adapts well to unseen languages and spontaneous speech. As a by-product, it can propose phoneme inventories for unseen languages. This is useful for language documentation.

Key points

  • Pre-training setup: supervised continual pre-training of HuBERT-base on 55 VoxCommunis languages, with targets from PanPhon articulatory features. The architecture adds a weighted layer sum and a BLSTM, following the SUPERB style.
  • Evaluation set: the ZRC2017 languages (English, French, German, Mandarin, Wolof), each with under 10 h of data, plus Swahili, Tamil, Thai, Turkish and Ukrainian.
  • Baselines: MMS, XEUS and mHuBERT-147. Adaptation to a new language uses only about 10 h of unlabelled speech, and Wolof works with under 2 h.
  • Zero-shot ABX error: on read speech, MauBERT-PHONE scores 5.02% against 5.12% for XEUS. On casual speech XEUS is ahead, at 9.99% against 10.58%.
  • Inventory discovery: this uses frequency thresholds on the feature vectors produced by MauBERT-FEAT.
    • A top-100 cutoff yields high recall (≥0.825 for four of five languages) but low precision (0.270–0.390).
    • Optimised thresholds raise precision to 0.778–0.872, with recall of 0.532–0.810.
    • Code is available at github.com/bootphon/maubert.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🔬Centre National de la Recherche ScientifiqueParis · France168 works · 62,369 citations🔬École Normale Supérieure - PSLParis · France23 works · 4,357 citations🎓Université Paris Sciences et LettresParis · France21 works · 3,290 citations🔬École des hautes études en sciences socialesParis · France15 works · 2,302 citations🎓Institut polytechnique de GrenobleGrenoble · France11 works · 207 citations🔬GIPSA-LabGrenoble · France3 works · 10 citations

 Fields, topics and concepts

📚 Shelves 7The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksPsychology · 1,552 worksExperimental and Cognitive Psychology · 870 worksSignal Processing · 452 works

🧭 Topics 1Research subjects and languages the work is about.Speech & spoken language · 480 works

Cite

Cite as Angelo Ortiz Tandazo, Manel Khentout, Youssef Benchekroun, Thomas Hueber and Emmanuel Dupoux. 2026. MauBERT: Universal Phonetic Inductive Biases for Few-Shot Acoustic Units Discovery. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.24

@inproceedings{tandazo2026maubert,
  title = {{MauBERT: Universal Phonetic Inductive Biases for Few-Shot Acoustic Units Discovery}},
  author = {Angelo Ortiz Tandazo, Manel Khentout, Youssef Benchekroun, Thomas Hueber and Emmanuel Dupoux},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.24},
  url = {https://doi.org/10.18653/v1/2026.acl-long.24}
}

References

60 references, 11 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Kanda, Takuya Yoshioka, Xiong Xiao, Jian Wu, Long Zhou, Shuo Ren, Yanmin Qian, Yao Qian, Jian Wu, Michael Zeng, Xiangzhan Yu, and Furu Wei. 2022. WavLM: Large-scale self-supervised pre-training for full stack speech processing. IEEE Journal of Selected Topics in Signal Processing, 16(6):1505– 1518. find
  2. Emily Ahn and Eleanor Chodroff. 2022. VoxCommunis: A corpus for cross-linguistic phonetic analysis. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 5286–5294, Marseille, France. European Language Resources Association. find
  3. Rosana Ardila, Megan Branson, Kelly Davis, Michael Kohler, Josh Meyer, Michael Henretty, Reuben Morais, Lindsay Saunders, Francis Tyers, and Gregor Weber. 2020. Common Voice: A massively-multilingual speech corpus. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 421…
  4. William Chen, Wangyou Zhang, Yifan Peng, Xinjian Li, Jinchuan Tian, Jiatong Shi, Xuankai Chang, Soumi Maiti, Karen Livescu, and Shinji Watanabe. 2024b. Towards robust speech representation learning for thousands of languages. In Proceedings of the 2024 Conference on Empirical Methods in Natural Lang… find
  5. Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, and Michael Auli. 2022. XLS-R: Self-supervised cross-lingual speech representation learning at scale. In Interspeech 2022…
  6. Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, and Michael Auli. 2021. Unsupervised cross-lingual representation learning for speech recognition. In Interspeech 2021, pages 2426– 2430. find
  7. Alexei Baevski, Yuhao Zhou, Abdelrahman Mohamed, and Michael Auli. 2020. wav2vec 2.0: A framework for self-supervised learning of speech representations. In Advances in Neural Information Processing Systems, volume 33, pages 12449–12460. Curran Associates, Inc.
  8. Can Balioglu, Alexander Erben, Martin Gleize, Artyom Kozhevnikov, Ilia Kulikov, and Julien Yao. 2023. fairseq2. find
  9. Zalán Borsos, Raphaël Marinier, Damien Vincent, Eugene Kharitonov, Olivier Pietquin, Matt Sharifi, Dominik Roblek, Olivier Teboul, David Grangier, Marco Tagliasacchi, and Neil Zeghidour. 2023. Audiolm: A language modeling approach to audio generation. IEEE/ACM Transactions on Audio, Speech, and Lang…
  10. Chih-Chen Chen, William Chen, Rodolfo Joel Zevallos, and John E Ortega. 2024a. Evaluating self-supervised speech representations for indigenous American languages. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2… find
  11. Peng-Jen Chen, Kevin Tran, Yilin Yang, Jingfei Du, Justine Kao, Yu-An Chung, Paden Tomasello, Paul-Ambroise Duquenne, Holger Schwenk, Hongyu Gong, Hirofumi Inaguma, Sravya Popuri, Changhan Wang, Juan Pino, Wei-Ning Hsu, and Ann Lee. 2023. Speech-to-speech translation for a real-world unwritten langu… find
  12. Sanyuan Chen, Chengyi Wang, Zhengyang Chen, Yu Wu, Shujie Liu, Zhuo Chen, Jinyu Li, Naoyuki Maureen de Seyssel, Marvin Lavechin, Yossi Adi, Emmanuel Dupoux, and Guillaume Wisniewski. 2022. Probing phoneme, language and speaker information in unsupervised speech representations. In Interspeech 2022,… find
  13. Alexandre Défossez, Laurent Mazaré, Manu Orsini, Amélie Royer, Patrick Pérez, Hervé Jégou, Edouard Grave, and Neil Zeghidour. 2024. Moshi: A speech-text foundation model for real-time dialogue. Preprint, arXiv:2410.00037. Version 2.
  14. Li Deng and Kevin Erler. 1991. Microstructural speech units and their hmm representation for discrete utterance speech recognition. In [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing, pages 193–196 vol.1. find
  15. Ewan Dunbar, Xuan Nga Cao, Juan Benjumea, Julien Karadayi, Mathieu Bernard, Laurent Besacier, Xavier Anguera, and Emmanuel Dupoux. 2017. The zero resource speech challenge 2017. In 2017 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU), pages 323–330. find
  16. Ewan Dunbar, Nicolas Hamilakis, and Emmanuel Dupoux. 2022. Self-supervised language learning from raw audio: Lessons from the zero resource speech challenge. IEEE Journal of Selected Topics in Signal Processing, 16(6):1211–1226. find
  17. Ellen Eide, Jan Robin Rohlicek, Herbert Gish, and Sanjoy Mitter. 1993. A linguistic feature representation of the speech waveform. In 1993 IEEE International Conference on Acoustics, Speech, and Signal Processing, volume 2, pages 483–486 vol.2. find
  18. Kjell Elenius and Gy Takacs. 1991. Phoneme recognition with an artificial neural network. In 2nd European Conference on Speech Communication and Technology (Eurospeech 1991), pages 121–124. find
  19. Siyuan Feng, Ming Tu, Rui Xia, Chuanzeng Huang, and Yuxuan Wang. 2023. Language-universal phonetic representation in multilingual speech pretraining for low-resource speech recognition. In Interspeech 2023, pages 1384–1388. find
  20. Mohamed, and Emmanuel Dupoux. 2021. On generative spoken language modeling from raw audio. Transactions of the Association for Computational Linguistics, 9:1336–1354.
  21. Joe Frankel, Mathew Magimai-Doss, Simon King, Karen Livescu, and Özgür Çetin. 2007. Articulatory feature classifiers trained on 2000 hours of telephone speech. In 8th Annual Conference of the International Speech Communication Association, INTERSPEECH 2007, Antwerp, Belgium, August 27-31, 2007, page… find
  22. Mark Hallap, Emmanuel Dupoux, and Ewan Dunbar. 2023. Evaluating context-invariance in unsupervised speech representations. In Interspeech 2023, pages 2973–2977. find
  23. Michael Hassid, Tal Remez, Tu Anh Nguyen, Itai Gat, Alexis Conneau, Felix Kreuk, Jade Copet, Alexandre Defossez, Gabriel Synnaeve, Emmanuel Dupoux, and 1 others. 2023. Textually pretrained speech language models. Advances in Neural Information Processing Systems, 36:63483–63501. find
  24. Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, and Abdelrahman Mohamed. 2021. HuBERT: Self-supervised speech representation learning by masked prediction of hidden units. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 29:3451–3460.
  25. J. Kahn, M. Rivière, W. Zheng, E. Kharitonov, Q. Xu, P.E. Mazaré, J. Karadayi, V. Liptchinsky, R. Collobert, C. Fuegen, T. Likhomanenko, G. Synnaeve, A. Joulin, A. Mohamed, and E. Dupoux. 2020. Libri-light: A benchmark for ASR with limited or no supervision. In ICASSP 2020 - 2020 IEEE International… find
  26. Simon King and Paul Taylor. 2000. Detection of phonological features in continuous speech using neural networks. Computer Speech & Language, 14(4):333– 353. find
  27. Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings.
  28. Marvin Lavechin, Maureen de Seyssel, Hadrien Titeux, Guillaume Wisniewski, Hervé Bredin, Alejandrina Cristia, and Emmanuel Dupoux. 2025. Simulating early phonetic and word learning without linguistic categories. Developmental Science, 28(2):e13606. find
  29. Alexander H. Liu, Heng-Jui Chang, Michael Auli, Wei-Ning Hsu, and Jim Glass. 2023. DinoSR: Self-Distillation and Online Clustering for Self-supervised Speech Representation Learning. Advances in Neural Information Processing Systems, 36:58346– 58362. find
  30. Karen Livescu, Ozgur Cetin, Mark Hasegawa-Johnson, Simon King, Chris Bartels, Nash Borges, Arthur Kantor, Partha Lal, Lisa Yung, Ari Bezman, Stephen Dawson-Haggerty, Bronwyn Woods, Joe Frankel, Mathew Magimai-Doss, and Kate Saenko. 2007. Articulatory feature-based methods for acoustic and audio-visu… find
  31. Ziyang Ma, Zhisheng Zheng, Guanrou Yang, Yu Wang, Chao Zhang, and Xie Chen. 2023. Pushing the limits of unsupervised unit discovery for SSL speech representation. In Interspeech 2023, pages 1269–1273. find
  32. Michael McAuliffe, Michaela Socolof, Sarah Mihuc, Michael Wagner, and Morgan Sonderegger. 2017. find
  33. Montreal Forced Aligner: Trainable Text-Speech Alignment Using Kaldi. In Proc. Interspeech 2017, pages 498–502. find
  34. Abdelrahman Mohamed, Hung-yi Lee, Lasse Borgholt, Jakob D. Havtorn, Joakim Edin, Christian Igel, Katrin Kirchhoff, Shang-Wen Li, Karen Livescu, Lars Maaløe, Tara N. Sainath, and Shinji Watanabe. 2022. Self-supervised speech representation learning: A review. IEEE Journal of Selected Topics in Signal… find
  35. Mukhtar Mohamed, Oli Danyi Liu, Hao Tang, and Sharon Goldwater. 2024. Orthogonality and isotropy of speaker and phonetic information in self-supervised speech representations. In Interspeech 2024, pages 3625–3629. find
  36. Katrin Kirchhoff. 1999. find
  37. David R. Mortensen, Patrick Littell, Akash Bharadwaj, Robust Speech Recognition Using Articulatory Information. Kartik Goyal, Chris Dyer, and Lori Levin. 2016. PhD dissertation, University of Bielefeld. PanPhon: A resource for mapping IPA segments to articulatory feature vectors. In Proceedings of P… find
  38. The COLING 2016 Springer Netherlands, Dordrecht. Organizing Committee. find
  39. Kushal Lakhotia, Eugene Kharitonov, Wei-Ning Hsu, Yossi Adi, Adam Polyak, Benjamin Bolte, Tu-Anh Nguyen, Jade Copet, Alexei Baevski, Abdelrahman Aaron van den Oord, Yazhe Li, and Oriol Vinyals. 2019. Representation learning with contrastive predictive coding. Preprint, arXiv:1807.03748. Version 2.
  40. George Papcun, Judith Hochberg, Timothy R. Thomas, François Laroche, Jeff Zacks, and Simon Levy. 1992. Inferring articulation and recognizing gestures from acoustics with a neural network trained on x-ray mi-crobeam data. The Journal of the Acoustical Society of America, 92(2):688–700. find
  41. Jiatong Shi, Dan Berrebbi, William Chen, En-Pei Hu, Wei-Ping Huang, Ho-Lam Chung, Xuankai Chang, Shang-Wen Li, Abdelrahman Mohamed, Hung-yi Lee, and Shinji Watanabe. 2023. ML-SUPERB: Multilingual speech universal performance benchmark. In Interspeech 2023, pages 884–888. find
  42. Fabian Pedregosa, Gaël Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Prettenhofer, Ron Weiss, Vincent Dubourg, Jake Vanderplas, Alexandre Passos, David Cournapeau, Matthieu Brucher, Matthieu Perrot, and Édouard Duchesnay. 2011. Scikit-learn:… find
  43. Marlene Staib, Tian Huey Teh, Alexandra Torresquin-tero, Devang S. Ram Mohan, Lorenzo Foglianti, Raphael Lenain, and Jiameng Gao. 2020. Phonological features for 0-shot multilingual speech synthesis. In Interspeech 2020, pages 2942–2946. find
  44. Maxime Poli, Emmanuel Chemla, and Emmanuel Dupoux. 2025a. fastabx: A library for efficient computation of abx discriminability. Preprint, arXiv:2505.02692. — [arXiv:2505.02692](https://arxiv.org/abs/2505.02692) find
  45. Maxime Poli, Manel Khentout, Angelo Ortiz Tandazo, Ewan Dunbar, Emmanuel Chemla, and Emmanuel Dupoux. 2026. Discophon: Benchmarking the unsupervised discovery of phoneme inventories with discrete speech units. Preprint, arXiv:2603.18612. — [arXiv:2603.18612](https://arxiv.org/abs/2603.18612) find
  46. Maxime Poli, Mahi Luthra, Youssef Benchekroun, Yosuke Higuchi, Martin Gleize, Jiayi Shen, Robin Algayres, Yu-An Chung, Mido Assran, Juan Pino, and Emmanuel Dupoux. 2025b. Spidr: Learning fast and stable linguistic units for spoken language models without supervision. Transactions on Machine Learning… find
  47. Vineel Pratap, Andros Tjandra, Bowen Shi, Paden Tomasello, Arun Babu, Sayani Kundu, Ali Elkahky, Zhaoheng Ni, Apoorv Vyas, Maryam Fazel-Zarandi, Alexei Baevski, Yossi Adi, Xiaohui Zhang, Wei-Ning Hsu, Alexis Conneau, and Michael Auli. 2024. Scaling speech technology to 1,000+ languages. Journal of M…
  48. Simon Rouard, Manu Orsini, Axel Roebel, Neil Zeghidour, and Alexandre Défossez. 2025. Continuous audio language models. Preprint, arXiv:2509.06926. Version 2. — [arXiv:2509.06926](https://arxiv.org/abs/2509.06926) find
  49. Thomas Schatz. 2016. ABX-Discriminability Measures and Applications. Theses, Université Paris 6 (UPMC). find
  50. Thomas Schatz, Naomi H. Feldman, Sharon Goldwater, Xuan-Nga Cao, and Emmanuel Dupoux. 2021. Early phonetic learning without phonetic categories: Insights from large-scale simulations on realistic input. Proceedings of the National Academy of Sciences, 118(7):e2001844118. find
  51. Thomas Schatz, Vijayaditya Peddinti, Francis Bach, Aren Jansen, Hynek Hermansky, and Emmanuel Dupoux. 2013. Evaluating speech features with the minimal-pair abx task: analysis of the classical mfc/plp pipeline. In Interspeech 2013, pages 1781– 1785. find
  52. Maarten Versteegh, Roland Thiollière, Thomas Schatz, Xuan Nga Cao, Xavier Anguera, Aren Jansen, and Emmanuel Dupoux. 2015. The zero resource speech challenge 2015. In Interspeech 2015, pages 3169– 3173. find
  53. Chengyi Wang, Yiming Wang, Yu Wu, Sanyuan Chen, Jinyu Li, Shujie Liu, and Furu Wei. 2022. Supervision-guided codebooks for masked prediction in speech pre-training. In Interspeech 2022, pages 2643–2647. find
  54. Chengyi Wang, Yu Wu, Yao Qian, Kenichi Kumatani, Shujie Liu, Furu Wei, Michael Zeng, and Xuedong Huang. 2021. Unispeech: Unified speech representation learning with labeled and unlabeled data. In Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Machi… find
  55. Janet F. Werker, Ferran Pons, Christiane Dietrich, Sachiyo Kajikawa, Laurel Fais, and Shigeaki Amano. 2007. Infant-directed speech supports phonetic category learning in English and Japanese. Cognition, 103(1):147–162. find
  56. Shu-wen Yang, Po-Han Chi, Yung-Sung Chuang, Cheng-I Jeff Lai, Kushal Lakhotia, Yist Y. Lin, Andy T. Liu, Jiatong Shi, Xuankai Chang, Guan-Ting Lin, Tzu-Hsien Huang, Wei-Cheng Tseng, Ko-tik Lee, Da-Rong Liu, Zili Huang, Shuyan Dong, Shang-Wen Li, Shinji Watanabe, Abdelrahman Mohamed, and Hung-yi Lee.… find
  57. Marcely Zanon Boito, Vivek Iyer, Nikolaos Lagos, Laurent Besacier, and Ioan Calapodescu. 2024. mHuBERT-147: A compact multilingual hubert model. In Interspeech 2024, pages 3939–3943. find
  58. Chen Zhang, Xu Tan, Yi Ren, Tao Qin, Kejun Zhang, and Tie-Yan Liu. 2021. Uwspeech: Speech to speech translation for unwritten languages. Proceedings of the AAAI Conference on Artificial Intelligence, 35(16):14319–14327. find
  59. Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen, Christopher Dewan, Mona Diab, Xian Li, Xi Victoria Lin, Todor Mihaylov, Myle Ott, Sam Shleifer, Kurt Shuster, Daniel Simig, Punit Singh Koura, Anjali Sridhar, Tianlu A Wang, and Luke Zettlemoyer. 2022. OPT: Open pre-tr…
  60. Piotr Żelasko, Siyuan Feng, Laureano Moro Velázquez, Ali Abavisani, Saurabhchand Bhati, Odette Scharenborg, Mark Hasegawa-Johnson, and Najim Dehak. 2022. Discovering phonetic inventories with crosslingual automatic speech recognition. Computer Speech & Language, 74:101358. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/tandazo-2026-maubert-universal-phonetic-inductive-biases/