Combinatorial Interaction Testing has been applied to event-driven software systems by using as test suite a set of sequences of inputs in desired combinations. This is generally called combinatorial sequence testing (CST). CST requires possibly new system models from which tests are generated and new test generation methods (or an adaptation of the classical ones). Finite State Machines (FSMs) can easily represent event-based systems where certain inputs are valid only in some states and such constraints can be represented by the incompleteness of the FSM. In this paper, we propose an approach to CST where tests are generated from FSMs which are represented by automata together with test requirements. First, automata can be used to check if test sequences contain invalid inputs. We propose three methods to repair tests with invalid inputs. Moreover, we can directly embed into automata the system constraints over the inputs during generations, to generate only valid test sequences. We compare our automata-based method with the standard approach of Sequences Covering Arrays (SCAs) that produces a set of sequences, all with the same length, composed by the permutation of all the events supported by the system. We found that generating only valid tests from automata provides several advantages iv.r.t. repairing tests and SCAs.