aiqbench analysis
HuPER represents a significant advancement in speech recognition by modeling phonetic perception as adaptive inference, achieving state-of-the-art results with minimal training data and strong zero-shot transfer capabilities across numerous languages. This framework's ability to handle diverse acoustic conditions through multi-path perception offers new insights into building more robust and efficient machine learning models for audio processing.
HuPER: A Human-Inspired Framework for Phonetic Perception
arXiv:2602.01634v2 Announce Type: replace-cross Abstract: 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/Berkeley-Speech-Group/HuPER.
Abstract: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 this https URL.
Audio and Speech Processing (eess.AS); Artificial Intelligence (cs.AI)
arXiv:2602.01634 [eess.AS]
(or arXiv:2602.01634v2 [eess.AS] for this version)
https://doi.org/10.48550/arXiv.2602.01634 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 173 words
Related coverage
A biological vision inspired framework for machine perception of abutting grating illusory contours
arXiv · Jun 16, 2026
The Machines that Make the Machines
NVIDIA · Oct 7, 2026
Human-Inspired Framework for Robotic Craniotomy: Integrating Multimodal Fusion and Adaptive Trajectory Adjustment
arXiv · Jul 24, 2026
Amazon Research Awards recipients announced
Amazon Science · May 27, 2026