Baidu Deep Speech system 81% accurate in noisy environments compared to 65% for best commercial systems
Arxiv – DeepSpeech: Scaling up end-to-end speech recognition Baidu researchers present a state-of-the-art speech recognition system developed using end-to-end deep learning. Our architecture is significantly simpler than traditional speech systems, which rely on laboriously engineered processing pipelines; these traditional systems also tend to perform poorly when used in noisy environments. In contrast, our system does …