Transformer-based segmentation of prosodic boundaries in Brazilian Portuguese

Researchers developed SAMPA, a Whisper-based segmenter that transcribes Brazilian Portuguese speech while inserting markers for terminal prosodic boundaries. Fine-tuned on the NURC-SP dataset, SAMPA achieved competitive F1 scores of 0.731 on held-out test data and 0.796 on the out-of-distribution MuPe-Diversidades dataset. The model follows morphosyntactic, semantic, and prosodic cues for boundary detection.

arXiv:2607.07408v1 Announce Type: new Abstract: Automatic prosodic segmentation identifies boundaries between speech units from acoustic and linguistic evidence. Although recent deep learning approaches have produced strong results for English, automatic segmentation for Brazilian Portuguese BP still relies mostly on rule-based or traditional machine-learning methods. This paper presents SAMPA, a Whisper-based segmenter that transcribes BP speech while inserting explicit markers for terminal prosodic boundaries. We fine-tune Whisper large-v3 on manually segmented recordings from the NURC-SP dataset and evaluate different training and test-time filtering configurations, including out-of-distribution testing on the MuPe-Diversidades dataset. SAMPA achieves competitive boundary-detection performance across settings, with the best models reaching F1=0.731 on the held-out test split and F1=0.796 on MuPe-Diversidades. Finally, through n-gram and acoustic-visual analyses, we show that our model follows morphosyntactic, semantic, and prosodic cues for detecting prosodic boundaries.