Speed controller plays an important role in high-performance induction motor vector control. Finite time control (FTC) is widely used in motor speed control because it makes the error converge to zero in a finite time. When the traditional FTC is far away from the convergence point, a larger control gain is needed to ensure a short convergence time since the speed error is large, which leads to decreased control efficiency, and sluggish error convergence rate, and the speed control performance degrades. Moreover, the finite time speed controller suffers from disturbances such as load change and parameter mismatch, which causes the controller unable to converge in finite time. Therefore, a finite time disturbance observer is employed for disturbance compensation of the proposed speed controller. Compared with the traditional finite time speed controller, the proposed speed control method has a better performance in rapidity and anti-disturbance. Simulation results verify the correctness and effectiveness of the proposed method.