Mice emit ultrasonic vocalizations (USVs) that are important for social communication. Despite great advancements in tools to detect USVs from audio files in recent years, highly accurate segmentation of USVs from spectrograms (i.e., removing noise) remains a significant challenge. Here, we present a new dataset of 12,954 annotated spectrograms explicitly labeled for mouse USV segmentation. Leveraging this dataset, we developed SqueakOut, a lightweight (4.6M parameters) fully convolutional autoencoder that achieves high accuracy in supervised segmentation of USVs from spectrograms, with a Dice score of 90.22. SqueakOut combines a MobileNetV2 backbone with skip connections and transposed convolutions to precisely segment USVs. Using stochastic data augmentation techniques and a hybrid loss function, SqueakOut learns robust segmentation across varying recording conditions. We evaluate SqueakOut's performance, demonstrating substantial improvements over existing methods like VocalMat (63.82 Dice score). The accurate USV segmentations enabled by SqueakOut will facilitate novel methods for vocalization classification and more accurate analysis of mouse communication. To promote further research, we release the annotated 12,954 spectrogram USV segmentation dataset and the SqueakOut implementation publicly.

中文翻译：

---

SqueakOut：基于自动编码器的小鼠超声发声分割

小鼠发出超声波发声（USV），这对于社交交流很重要。尽管近年来从音频文件中检测 USV 的工具取得了巨大进步，但从频谱图中对 USV 进行高精度分割（即消除噪声）仍然是一个重大挑战。在这里，我们提出了一个新的数据集，其中包含 12,954 个带注释的频谱图，这些频谱图明确标记为小鼠 USV 分割。利用该数据集，我们开发了 SqueakOut，这是一种轻量级（4.6M 参数）全卷积自动编码器，可在声谱图中对 USV 进行监督分割，实现高精度，Dice 得分为 90.22。 SqueakOut 将 MobileNetV2 主干与跳跃连接和转置卷积相结合，以精确分割 USV。使用随机数据增强技术和混合损失函数，SqueakOut 可以在不同的记录条件下学习稳健的分割。我们评估了 SqueakOut 的性能，证明其比 VocalMat（63.82 Dice 得分）等现有方法有显着改进。 SqueakOut 实现的精确 USV 分割将促进发声分类的新方法和更准确的鼠标通信分析。为了促进进一步的研究，我们公开发布了带注释的 12,954 个频谱图 USV 分割数据集和 SqueakOut 实现。