Lessons learned in part-of-speech tagging of conversational speech

This paper examines tagging models for spontaneous English speech transcripts. We analyze the performance of state-of-the-art tagging models, either generative or discriminative, left-to-right or bidirectional, with or without latent annotations, together with the use of ToBI break indexes and several methods for segmenting the speech transcripts (i.e., conversation side, speaker turn, or human-annotated sentence). Based on these studies, we observe that: (1) bidirectional models tend to achieve better accuracy levels than left-to-right models, (2) generative models seem to perform somewhat better than discriminative models on this task, and (3) prosody improves tagging performance of models on conversation sides, but has much less impact on smaller segments. We conclude that, although the use of break indexes can indeed significantly improve performance over baseline models without them on conversation sides, tagging accuracy improves more by using smaller segments, for which the impact of the break indexes is marginal.