Which Prosodic Features Matter Most for Pragmatics?
This work addresses the problem of identifying key prosodic features for pragmatics, which can guide basic research in prosody and improve applications like speech synthesis evaluation, though it is incremental in nature.
The study investigated which prosodic features are most important for conveying pragmatic functions by predicting human perceptions of pragmatic similarity in utterance pairs. It found that duration-related features are more important than pitch-related ones, and utterance-initial features matter more than utterance-final ones, with neglected features like nasality and vibrato also showing significance.
We investigate which prosodic features matter most in conveying prosodic functions. We use the problem of predicting human perceptions of pragmatic similarity among utterance pairs to evaluate the utility of prosodic features of different types. We find, for example, that duration-related features are more important than pitch-related features, and that utterance-initial features are more important than utterance-final features. Further, failure analysis indicates that modeling using pitch features only often fails to handle important pragmatic functions, and suggests that several generally-neglected acoustic and prosodic features are pragmatically significant, including nasality and vibrato. These findings can guide future basic research in prosody, and suggest how to improve speech synthesis evaluation, among other applications.