Fig. 1From: Adversarial joint training with self-attention mechanism for robust end-to-end speech recognitionOverview of the SE_ASR joint training frameworkBack to article page