Different approaches of minima tracking based noise estimation algorithms are compared and modifications increasing their efficiency are proposed. Estimated noise is used by noise suppression algorithm that is a part of speech recognition system. Moreover, the algorithms are developed to be applied in feature extraction of Distributed Speech Recognition (DSR). Therefore we propose such modifications to the noise estimation techniques that are quickly adaptable on varying noise and do not need so much information from past segments. We also minimized the algorithmic delay. The robustness of proposed algorithms were tested under several noisy conditions on five Speech-Dat Car (SDC) and Aurora 2 evaluation databases.