None defined yet.
Towards Quantifying Benchmark Optimization in ASR Models
TADA: A Generative Framework for Speech Modeling via Text-Acoustic Dual Alignment