Research in ASR (Automatic Speech Recognition) aims to enable computers to "understand" human speech and convert it into text.