SURE is a new standardized framework for evaluating and training speech foundation models and Speech LLMs to improve comparability and reproducibility under realistic conditions.
Contextasr-bench: A massive contextual speech recognition benchmark
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
dataset 1
citation-polarity summary
years
2026 3roles
dataset 1polarities
use dataset 1representative citing papers
A survey of Large Audio Language Models that establishes a taxonomy of trustworthiness vulnerabilities and proposes a Defense-in-Depth roadmap for audio intelligence.
citing papers explorer
No citing papers match the current filters.