Gary Wang

data augmentation (5) automatic speech recognition (4) text injection (2) semi-supervised learning (2) speech recognition (2) text-to-speech synthesis (2) connectionist temporal classification (1) generative adversarial network (1) parameter efficiency (1) word error rate (1) consistency training (1) non-autoregressive model (1) recurrent neural network transducer (1) speaker diversity (1) noisy student training (1) acoustic diversity (1) speech recognition robustness (1) personal identifiable information (1) contrastive learning (1) non-parallel voice conversion (1)