The time tabling problems in general are NP-complete. University course timetabling problem involves in assignment of a set of courses to a given number of periods and rooms subject to many hard and soft constraints. Hard constraints must be satisfied to producea feasible solution. Soft constraints should be satisfied as much as possible. The manual solution of a real-world timetabling problem usually requires several days or even weeks. Therefore, many research has been invested to automated timetabling. In the timetabling literature, there are many meta-heuristic approaches for solving these problems include Tabu Search, Simulated Annealing, Ant Colony Optimization, Memetic Algorithm, Genetic Agorithm, Harmony Search, Local Search, etc. According to a report of ITC07 the International Timetabling Competition 2007, the university course timetabling problem can be classified into two groups: curriculum-based course timetabling problems and post enrolment-based course timetabling problems. The curriculum-based course timetabling problem is very popular in many universities in Vietnam. In this paper, hybrid GA – Bees algorithm for this problem is presented. The hybrid algorithm combines the natural evolution property (such as inheritance, selection, mutation and crossover) of Genetic algorithm and the particular ability of the Bees algorithm to avoid local optima. The approach is tested over 14 real world data which is taken from the Faculty of Information Technology, HCMC University of Science in Vietnam and compared with previous works. The results demonstrate that our approach is able to produce high quality solutions against others that appear in the literature.