Previous works have shown that remarkable performance improvements can be attained in speaker and language recognition tasks by combining several heterogeneous systems that provide complementary information. In this work, the complementarity of several i-vector language recognition systems, using Mel-Frequency Cepstral-Coefficient (MFCC) features computed on Short-Time Fourier Analysis windows of different sizes, is studied. Language recognition experiments carried out on the NIST 2007 and 2009 LRE datasets reveal relative performance gains of up to 33% when fusing the systems, with regard to the best single system. Results suggest that combining acoustic systems based on analysis windows of different sizes may allow to get advantage from both the sharper characterization of short events provided by short windows and the better frequency resolution of stationary events provided by long windows.