This paper deals with the problem concerned with output-feedback adaptive neural control for a class of stochastic nonlinear time-delay systems with unknown control directions. Firstly, the state observer is established for estimating the unmeasured states. Then, in the design procedure, an appropriate Lyapunov-Krasovskii functional is used to compensate the unknown time-delay terms, the Nussbaum-type gain function is used to handle the unknown control directions, the neural network is employed to approximate the unknown nonlinearities, and the dynamic surface control(DSC) technique is used to avoid the the problem of éxplosion of complexity' in control design. Finally, a robust adaptive neural output feedback control scheme is constructed via the Lyapunov stability theory. It is shown that the designed controller can ensure that all the signals in the closed-loop system are 4-Moment (or 2 Moment) semi-globally uniformly ultimately bounded(SGUUB) and the tracking error converges to a small neighborhood of the origin. The example is presented to verify the effectiveness of the proposed scheme.