Papers by Mansi Dhamne
Predicting Prosodic Boundaries for Children’s Texts (2025.emnlp-main)
Copied to clipboard
| Challenge: | Using a dataset of 54 leveled English stories annotated for potential pauses, we find that nearly 30% of pause occur at non-punctuation locations of the text. |
| Approach: | They propose to use a text-based model to predict pause locations in children's reading material using a curated dataset of 54 leveled English stories annotated for potential pauses, or prosodic boundaries, by 21 fluent speakers. |
| Outcome: | The proposed model can model both allowed and “forbidden” pauses . it uses a curated dataset of 54 leveled English stories annotated for potential pause locations by 21 fluent speakers . |