The role of context in neural pitch accent detection in English
This work addresses the problem of detecting prosodic events for downstream NLP tasks, but it is incremental as it builds on prior CNN-based methods.
The authors tackled pitch accent detection in English speech by proposing a model that uses full utterances and an LSTM layer for better context, achieving an improvement from 87.5% to 88.7% accuracy on the Boston University Radio News Corpus.
Prosody is a rich information source in natural language, serving as a marker for phenomena such as contrast. In order to make this information available to downstream tasks, we need a way to detect prosodic events in speech. We propose a new model for pitch accent detection, inspired by the work of Stehwien et al. (2018), who presented a CNN-based model for this task. Our model makes greater use of context by using full utterances as input and adding an LSTM layer. We find that these innovations lead to an improvement from 87.5% to 88.7% accuracy on pitch accent detection on American English speech in the Boston University Radio News Corpus, a state-of-the-art result. We also find that a simple baseline that just predicts a pitch accent on every content word yields 82.2% accuracy, and we suggest that this is the appropriate baseline for this task. Finally, we conduct ablation tests that show pitch is the most important acoustic feature for this task and this corpus.