Recognition of helicopter by acoustic signal is an important way to identify helicopter, and the key is feature extraction. Inspired by the fact that the human auditory system can stably recognize target in the low signal-to-noise ratio and complex environments, the auditory computing model is introduced into the traditional spectrum feature extraction. A unified framework for recognizing helicopter acoustic signal based on auditory spectrum feature extraction is proposed and six specific algorithms under this framework are given. Numerical simulation results validated the effectiveness and robustness of the proposed algorithms. The results showed that the recognition rate and noise robustness of the proposed feature extraction algorithms are higher when the data segment length and the number of the auditory filters are properly set such that the frequency resolution is of the same order as the fundamental frequency of the helicopter. The results also showed that the lower bound of the analyzed frequency range has greater influence on the recognition performance than the upper bound. In order to improve the recognition performance, computational models, which are closer to the real human auditory perception and can enhance the resolution of the scale transformation in the low-to-medium frequency and the sharpness of frequency selection of the auditory filter bank, are desired.