Generation of human auditory steady-state responses