Employing self-supervised learning models for cross-linguistic child speech maturity classification
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916788544995328 |
|---|---|
| author | Zhang, Theo Suresh, Madurya Warlaumont, Anne S. Hitczenko, Kasia Cristia, Alejandrina Cychosz, Margaret |
| author_facet | Zhang, Theo Suresh, Madurya Warlaumont, Anne S. Hitczenko, Kasia Cristia, Alejandrina Cychosz, Margaret |
| contents | Speech technology systems struggle with many downstream tasks for child speech due to small training corpora and the difficulties that child speech pose. We apply a novel dataset, SpeechMaturity, to state-of-the-art transformer models to address a fundamental classification task: identifying child vocalizations. Unlike previous corpora, our dataset captures maximally ecologically-valid child vocalizations across an unprecedented sample, comprising children acquiring 25+ languages in the U.S., Bolivia, Vanuatu, Papua New Guinea, Solomon Islands, and France. The dataset contains 242,004 labeled vocalizations, magnitudes larger than previous work. Models were trained to distinguish between cry, laughter, mature (consonant+vowel), and immature speech (just consonant or vowel). Models trained on the dataset outperform state-of-the-art models trained on previous datasets, achieved classification accuracy comparable to humans, and were robust across rural and urban settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_08999 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Employing self-supervised learning models for cross-linguistic child speech maturity classification Zhang, Theo Suresh, Madurya Warlaumont, Anne S. Hitczenko, Kasia Cristia, Alejandrina Cychosz, Margaret Computation and Language Artificial Intelligence Speech technology systems struggle with many downstream tasks for child speech due to small training corpora and the difficulties that child speech pose. We apply a novel dataset, SpeechMaturity, to state-of-the-art transformer models to address a fundamental classification task: identifying child vocalizations. Unlike previous corpora, our dataset captures maximally ecologically-valid child vocalizations across an unprecedented sample, comprising children acquiring 25+ languages in the U.S., Bolivia, Vanuatu, Papua New Guinea, Solomon Islands, and France. The dataset contains 242,004 labeled vocalizations, magnitudes larger than previous work. Models were trained to distinguish between cry, laughter, mature (consonant+vowel), and immature speech (just consonant or vowel). Models trained on the dataset outperform state-of-the-art models trained on previous datasets, achieved classification accuracy comparable to humans, and were robust across rural and urban settings. |
| title | Employing self-supervised learning models for cross-linguistic child speech maturity classification |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2506.08999 |