语音识别(ASR)论文优选：A comparison of streaming models and data augmentation methods for robust speech recog
本文对比端到端流式ASR系统Monotonic Chunkwise Attention (MoChA) 和 Recurrent Neural Network-Transducer (RNN-T)的性能以及acoustic simulator, Vocal Tract Length Perturbation (VTLP) for speaker variability, and SpecAugment等数据增广对比试验，主要为经验分享。
