Improvement in Automatic Pronunciation Scoring using Additional Basic Scores and Learning to Rank

This paper proposes the adoption of different word-level scores in the framework of automatic pronunciation scoring using learning to rank. Six types of phone-level scores are first computed and converted to word-level scores by using average-based, vowel-based, and consonant-based methods. Different score combination methods are then used to combine these word-level scores to obtain the final combined score for the utterance under inspection. The experimental result shows that the learning to rank methods perform better than most of the existing methods, while using all types of word-level scores can lead to some improvement over the original average-based scores.