Fast and precise beam alignment is crucial for high-quality data transmission in millimeter-wave (mmWave) communication systems, where large-scale antenna arrays are utilized to overcome the severe propagation loss. To tackle the challenging problem, we propose a novel deep learning-based hierarchical beam alignment method for both multiple-input single-output (MISO) and multiple-input multiple-output (MIMO) systems, which learns two tiers of probing codebooks (PCs) and uses their measurements to predict the optimal beam in a coarse-to-fine search manner. Specifically, a hierarchical beam alignment network (HBAN) is developed for MISO systems, which first performs coarse channel measurement using a tier-1 PC, then selects a tier-2 PC for fine channel measurement, and finally predicts the optimal beam based on both coarse and fine measurements. The propounded HBAN is trained in two steps: the tier-1 PC and the tier-2 PC selector are first trained jointly, followed by the joint training of all the tier-2 PCs and beam predictors. Furthermore, an HBAN for MIMO systems is proposed to directly predict the optimal beam pair without performing beam alignment individually at the transmitter and receiver. Numerical results demonstrate that the proposed HBANs are superior to the state-of-art methods in both alignment accuracy and signaling overhead reduction.