Figure 1From: Optimizing Automatic Speech Recognition for Low-Proficient Non-Native Speakers System architecture. Back to article page