The processes observed in a sound environment inevitably contain additional external noise of arbitrary distribution. Furthermore, the actual sound environment system exhibits various types of linear and non-linear characteristics, and it often contains an unknown structure. In this paper, a method for estimating the input signal for a sound environment system with unknown structure and additive noise of arbitrary probability distribution is proposed by introducing a system model of the conditional probability type. The effectiveness of the proposed theoretical method is confirmed experimentally by applying it to the actual problem of input estimation of the sound environment.