For visually impaired people, auditory information is one of the most important sources of information for orientation and mobility. Therefore, when we design acoustic environments, we must not ignore their actual demands concerning sounds. Several studies revealed how the visually impaired use the sounds around them have been conducted. In these studies, the usage of sounds of visually impaired people was revealed qualitatively, but acoustical properties of ideal designs for them were not known sufficiently, even a basic property such as adequate sound level of an acoustic traffic signal. In this study, psycho-acoustical experiments were conducted to examine the methods to determine adequate sound levels of acoustical "signs" (auditory signals, announcements and advertisements) for the visually impaired. Environmental sounds were recorded with HATS (Head and Torso Simulator) and a stereo microphone simultaneously. These recorded environmental sounds were presented to participants at the recorded sound level. Several acoustical signs that visually impaired often use for orientation and mobility were also recorded with HATS and a stereo microphone. Each recorded environmental sound and acoustical sign were played simultaneously. The participants were required to adjust the level of each acoustical sign at their adequate level comparing with the environmental sound. The result was that the adjusted levels of participants for each recording condition were similar to each other. This suggests that we can determine the adequate sound level of acoustical signs for the visually impaired using sounds recorded with HATS via headphones. The visually impaired with poor walking ability, whose opinions have not been heard appropriately, can also participate in the experiments.