SAMPA：基于 Transformer 的巴西葡萄牙语韵律边界分割
SAMPA: Transformer-based segmentation of prosodic boundaries in Brazilian Portuguese
AI 导读
SAMPA 是一个基于 Whisper large-v3 微调的巴西葡萄牙语韵律边界分割器，在转写语音时插入显式的终端韵律边界标记，训练数据为 NURC-Sampa 数据集的手工标注录音。模型在留出测试集上达到 f1 binary 0.730，在分布外的 MuPe-Diversidades 数据集上达到 0.793，优于开箱即用的 Whisper 基线和重训练的特征分类器。
正文 · 原文
Abstract:Automatic prosodic segmentation identifies boundaries between speech units based on acoustic and linguistic evidence. Although recent deep learning approaches have achieved strong results for English, automatic segmentation for Brazilian Portuguese (BP) still relies largely 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-Sampa dataset and evaluate different training and test-time filtering configurations, including out-of-distribution testing on the MuPe-Diversidades dataset. Without test-time filtering, the best models reach f1 binary of 0.730 on the held-out test split and 0.793 on MuPe-Diversidades, outperforming an out-of-the-box Whisper baseline and a retrained feature-based classifier. Finally, n-gram and acoustic-visual analyses provide qualitative observations that help interpret the model's predictions and formulate hypotheses about the linguistic and prosodic cues that may influence its boundary decisions.
| Comments: | 11 pages, 2 figures, submitted to an international conference |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.07408 [cs.CL] |
| (or arXiv:2607.07408v3 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.07408 arXiv-issued DOI via DataCite |
Submission history
From: Rodrigo De Freitas Lima [view email]
[v1]
Wed, 8 Jul 2026 13:41:43 UTC (886 KB)
[v2]
Mon, 13 Jul 2026 16:03:28 UTC (886 KB)
[v3]
Fri, 23 Oct 2026 21:39:37 UTC (838 KB)
来源：arXiv cs.AI · arxiv.org