Askar Hamdulla

speaker recognition (2) self-supervised learning (1) data augmentation (1) speech recognition (1) keyword spotting (1) saliency map (1) convolutional neural network (1) model interpretation (1) class activation map (1) second language acquisition (1) speaker identification (1) speech corpus (1) pronunciation error (1) syllable labeling (1) tone labeling (1) chinese l2 learning (1) neural network (1) few-shot learning (1) mixed speech (1)