Sensory substitution devices (SSDs) provide visually impaired users with a 'visual' perception of the environment by transmitting information about the field of view of the head-mounted camera. To increase the effectiveness of SSDs, the visual data should be pre-processed to convey the spatial information about the object in simple and understandable way. In this work, the authors supposed that the use of the stereo approach would give the user more information about the environment and increase the accuracy in spatial localization task. The experiment involved 30 sighted participants, who were blindfolded during the experiment. 15 participants were tested in mono mode and 15 participants were tested in stereo mode. For both modes, positive dynamics in spatial localization accuracy was observed during the experiment. The absolute deviation and the X-axis deviation (in width) were significantly higher in stereo mode. No significant difference was found in Y-axis deviation (in depth). The authors suggest that the accuracy in the stereo mode may be improved by increasing the time of training with the SSD.