We propose HuPER, a human-inspired framework that models phonetic perception as adaptive inference over acoustic-phonetics evidence and linguistic knowledge. With only 100 hours of training data, HuPER achieves state-of-the-art phonetic error rates on five English benchmarks and strong zero-shot transfer to 95 unseen languages. HuPER is also the first framework to enable adaptive, multi-path phonetic perception under diverse acoustic conditions. All training data, models, and code are open-sourced. Code and demo avaliable at https://github.com/HuPER29/HuPER.
@article{arxiv.2602.01634,
title = {HuPER: A Human-Inspired Framework for Phonetic Perception},
author = {Chenxu Guo and Jiachen Lian and Yisi Liu and Baihe Huang and Shriyaa Narayanan and Cheol Jun Cho and Gopala Anumanchipalli},
journal= {arXiv preprint arXiv:2602.01634},
year = {2026}
}