Abstract: This study investigated the effects of perceived spatial separation induced by precedence effect on Chinese speech recognition with informational masking. In the experiment, two loudspeakers which located in left and right respectively were used to play the target and masking, and the delay between the two loudspeakers was changed to produce different perceived spatial location of the masking. The results suggest that although the target and masking are sent from the same loudspeakers, the perceived spatial separation can improve the speech recognition. Under informational masking, the improvements are significant greater than under energetic masking. These results are useful in the separation of energetic and informational masking of speech.