In this paper, the transonic measurement of air velocity is addressed based on an acoustic vector sensor. According to the measurement model of acoustic particle-velocity vector field developed by the analytical method of the equivalent acoustic source, we establish the relationship between the observation data of the acoustic vector sensor and the transonic air velocity to be estimated. A two-dimensional Generalized Multiple Signal Classification algorithm for Air Velocity Estimation (GMUSIC-AVE) is proposed. The GMUSIC-AVE algorithm can not only make the correct judgment between the subsonic and supersonic air velocity, but also yield high estimation accuracy of the current air velocity. Simulations are implemented to illustrate the effectiveness of the proposed algorithm.