We calculate the conversion from non-adiabatic, non-radial oscillations tidally induced by a hot Jupiter on a star to observable spectroscopic and photometric signals. Models with both frozen convection and an approximation for a perturbation to the convective flux are discussed. Observables are calculated for some real planetary systems to give specific predictions. Time-dependent line broadening and the radial velocity signal during transit are both investigated as methods to provide further insight into the nature of the stellar oscillations. The photometric signal is predicted to be proportional to the inverse square of the orbital period, P −2 , as in the equilibrium tide approximation. However, the radial velocity signal is predicted to be proportional to P −1 , and is therefore much larger at long orbital periods than the signal corresponding to the equilibrium tide approximation, which is proportional to P −3. The prospects for detecting these oscillations and the implications for the detection and characterisation of planets are discussed.