Health Acoustic Representations (HeAR) is a bioacoustic foundation model developed by the Google Research team, aimed at identifying early signs of disease by analyzing sounds produced by the human body, such as coughs. The model has been trained on 300 million audio samples, utilizing approximately 100 million samples specifically for cough sounds. HeAR can recognize health-related sound patterns, providing a robust foundation for medical audio analysis. The HeAR model outperforms other models across various tasks and demonstrates superior generalization capabilities across different microphones. Additionally, HeAR allows models trained with limited data to achieve high performance, which is crucial in data-scarce medical research fields. HeAR is currently open to researchers to accelerate the development of customized bioacoustic models while reducing the need for data, setup, and computing resources.