Leveraging phone-level linguistic-acoustic similarity for utterance-level pronunciation scoring

Re-assign community