Cardiovascular diseases are the leading cause of death worldwide, with an estimate of 18 million deaths per year globally. To address this important public health problem and develop novel therapies it is necessary to further our understanding of the cellular composition of the heart. Here we use large scale single cell omics to sample 500, 000 cells from six major cardiac regions in 14 healthy female and male donors. Using cutting-edge machine learning technologies, we provide an integrative single cell atlas of the human heart, revealing high levels of cellular heterogeneity in different anatomical regions, previously unknown cardiac cell states and their transcriptional circuits, and how these change in disease.