Convex Low-resource Accent-Robust Language Detection in Speech Recognition

Miria Feng, William Tan, Mert Pilanci
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:30379-30399, 2026.

Abstract

Globalization and multiculturalism continue to produce increasingly diverse speech varieties. Yet current spoken dialogue systems frequently fail on under-represented dialects and accents, often misidentifying the input language and causing cascading failures in downstream dialogue tasks. Addressing this dialectal variance under low-resource constraints remains an open challenge, as standard fine-tuning is computationally expensive and prone to overfitting on high-dimensional speech data. We propose Convex Language Detection (CLD), a novel framework that integrates theoretically grounded convex optimization techniques into the spoken dialogue systems pipeline. Our method is efficiently implemented via multi-GPU Alternating Direction Method of Multipliers (ADMM) in JAX, thus providing global optimality guarantees and fast training in polynomial time. Theoretically, we prove that our convex objective induces certified margin stability and provide guarantees against feature perturbations. Empirically, we demonstrate sample efficiency and robustness to input dialectical variation, achieving 97–98% accuracy in challenging low-resource regimes. Our open-source package is available at https://pypi.org/project/jaxcld/.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-feng26p, title = {Convex Low-resource Accent-Robust Language Detection in Speech Recognition}, author = {Feng, Miria and Tan, William and Pilanci, Mert}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {30379--30399}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/feng26p/feng26p.pdf}, url = {https://proceedings.mlr.press/v306/feng26p.html}, abstract = {Globalization and multiculturalism continue to produce increasingly diverse speech varieties. Yet current spoken dialogue systems frequently fail on under-represented dialects and accents, often misidentifying the input language and causing cascading failures in downstream dialogue tasks. Addressing this dialectal variance under low-resource constraints remains an open challenge, as standard fine-tuning is computationally expensive and prone to overfitting on high-dimensional speech data. We propose Convex Language Detection (CLD), a novel framework that integrates theoretically grounded convex optimization techniques into the spoken dialogue systems pipeline. Our method is efficiently implemented via multi-GPU Alternating Direction Method of Multipliers (ADMM) in JAX, thus providing global optimality guarantees and fast training in polynomial time. Theoretically, we prove that our convex objective induces certified margin stability and provide guarantees against feature perturbations. Empirically, we demonstrate sample efficiency and robustness to input dialectical variation, achieving 97–98% accuracy in challenging low-resource regimes. Our open-source package is available at https://pypi.org/project/jaxcld/.} }
Endnote
%0 Conference Paper %T Convex Low-resource Accent-Robust Language Detection in Speech Recognition %A Miria Feng %A William Tan %A Mert Pilanci %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-feng26p %I PMLR %P 30379--30399 %U https://proceedings.mlr.press/v306/feng26p.html %V 306 %X Globalization and multiculturalism continue to produce increasingly diverse speech varieties. Yet current spoken dialogue systems frequently fail on under-represented dialects and accents, often misidentifying the input language and causing cascading failures in downstream dialogue tasks. Addressing this dialectal variance under low-resource constraints remains an open challenge, as standard fine-tuning is computationally expensive and prone to overfitting on high-dimensional speech data. We propose Convex Language Detection (CLD), a novel framework that integrates theoretically grounded convex optimization techniques into the spoken dialogue systems pipeline. Our method is efficiently implemented via multi-GPU Alternating Direction Method of Multipliers (ADMM) in JAX, thus providing global optimality guarantees and fast training in polynomial time. Theoretically, we prove that our convex objective induces certified margin stability and provide guarantees against feature perturbations. Empirically, we demonstrate sample efficiency and robustness to input dialectical variation, achieving 97–98% accuracy in challenging low-resource regimes. Our open-source package is available at https://pypi.org/project/jaxcld/.
APA
Feng, M., Tan, W. & Pilanci, M.. (2026). Convex Low-resource Accent-Robust Language Detection in Speech Recognition. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:30379-30399 Available from https://proceedings.mlr.press/v306/feng26p.html.

Related Material