HuPER: A Human-Inspired Framework for Phonetic Perception

TL;DR

HuPER is a novel human-inspired framework that models phonetic perception as adaptive inference, achieving state-of-the-art results with limited data and strong zero-shot transfer across many languages.

Contribution

It introduces HuPER, the first framework enabling adaptive, multi-path phonetic perception under diverse acoustic conditions with minimal training data.

Findings

Achieves state-of-the-art phonetic error rates on five English benchmarks.

Demonstrates strong zero-shot transfer to 95 unseen languages.

Enables adaptive perception under diverse acoustic environments.

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/HuPER29/HuPER.