Phil Woodland

large language model (2) simultaneous speech translation (2) speech recognition (2) uncertainty quantification (1) speech synthesis (1) end-to-end training (1) weakly supervised learning (1) out-of-domain detection (1) multitask learning (1) streaming inference (1) parameter-efficient fine-tuning (1) low-rank adaptation (1) evidential deep learning (1) reference-free evaluation (1) discrete token (1) automatic speech recognition (1) continuous representation (1) speech-to-speech translation (1) end-to-end translation (1) self-supervised learning (1)