InfantCryNet: A Data-driven Framework for Intelligent Analysis of Infant Cries

Mengze Hong, Chen Jason Zhang, Lingxiao Yang, Yuanfeng SONG, Di Jiang
Proceedings of the 16th Asian Conference on Machine Learning, PMLR 260:845-857, 2025.

Abstract

Understanding the meaning of infant cries is a significant challenge for young parents in caring for their newborns. The presence of background noise and the lack of labeled data present practical challenges in developing systems that can detect crying and analyze its underlying reasons. In this paper, we present a novel data-driven framework, “InfantCryNet,” for accomplishing these tasks. To address the issue of data scarcity, we employ pre-trained audio models to incorporate prior knowledge into our model. We propose the use of statistical pooling and multi-head attention pooling techniques to extract features more effectively. Additionally, knowledge distillation and model quantization are applied to enhance model efficiency and reduce the model size, better supporting industrial deployment in mobile devices. Experiments on real-life datasets demonstrate the superior performance of the proposed framework, outperforming state-of-the-art baselines by 4.4% in classification accuracy. The model compression effectively reduces the model size by 7% without compromising performance and by up to 28% with only an 8% decrease in accuracy, offering practical insights for model selection and system design.

Cite this Paper

BibTeX
@InProceedings{pmlr-v260-hong25a, title = {{InfantCryNet}: {A} Data-driven Framework for Intelligent Analysis of Infant Cries}, author = {Hong, Mengze and Zhang, Chen Jason and Yang, Lingxiao and SONG, Yuanfeng and Jiang, Di}, booktitle = {Proceedings of the 16th Asian Conference on Machine Learning}, pages = {845--857}, year = {2025}, editor = {Nguyen, Vu and Lin, Hsuan-Tien}, volume = {260}, series = {Proceedings of Machine Learning Research}, month = {05--08 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v260/main/assets/hong25a/hong25a.pdf}, url = {https://proceedings.mlr.press/v260/hong25a.html}, abstract = {Understanding the meaning of infant cries is a significant challenge for young parents in caring for their newborns. The presence of background noise and the lack of labeled data present practical challenges in developing systems that can detect crying and analyze its underlying reasons. In this paper, we present a novel data-driven framework, “InfantCryNet,” for accomplishing these tasks. To address the issue of data scarcity, we employ pre-trained audio models to incorporate prior knowledge into our model. We propose the use of statistical pooling and multi-head attention pooling techniques to extract features more effectively. Additionally, knowledge distillation and model quantization are applied to enhance model efficiency and reduce the model size, better supporting industrial deployment in mobile devices. Experiments on real-life datasets demonstrate the superior performance of the proposed framework, outperforming state-of-the-art baselines by 4.4% in classification accuracy. The model compression effectively reduces the model size by 7% without compromising performance and by up to 28% with only an 8% decrease in accuracy, offering practical insights for model selection and system design.} }
Endnote
%0 Conference Paper %T InfantCryNet: A Data-driven Framework for Intelligent Analysis of Infant Cries %A Mengze Hong %A Chen Jason Zhang %A Lingxiao Yang %A Yuanfeng SONG %A Di Jiang %B Proceedings of the 16th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Vu Nguyen %E Hsuan-Tien Lin %F pmlr-v260-hong25a %I PMLR %P 845--857 %U https://proceedings.mlr.press/v260/hong25a.html %V 260 %X Understanding the meaning of infant cries is a significant challenge for young parents in caring for their newborns. The presence of background noise and the lack of labeled data present practical challenges in developing systems that can detect crying and analyze its underlying reasons. In this paper, we present a novel data-driven framework, “InfantCryNet,” for accomplishing these tasks. To address the issue of data scarcity, we employ pre-trained audio models to incorporate prior knowledge into our model. We propose the use of statistical pooling and multi-head attention pooling techniques to extract features more effectively. Additionally, knowledge distillation and model quantization are applied to enhance model efficiency and reduce the model size, better supporting industrial deployment in mobile devices. Experiments on real-life datasets demonstrate the superior performance of the proposed framework, outperforming state-of-the-art baselines by 4.4% in classification accuracy. The model compression effectively reduces the model size by 7% without compromising performance and by up to 28% with only an 8% decrease in accuracy, offering practical insights for model selection and system design.
APA
Hong, M., Zhang, C.J., Yang, L., SONG, Y. & Jiang, D.. (2025). InfantCryNet: A Data-driven Framework for Intelligent Analysis of Infant Cries. Proceedings of the 16th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 260:845-857 Available from https://proceedings.mlr.press/v260/hong25a.html.

Related Material