A method and an apparatus for detecting an audio signal according to frequency domain energy is presented. The method may include receiving an audio signal frame; acquiring frequency domain energy distribution of the audio signal frame; obtaining a maximum value distribution characteristic of a frequency domain energy distribution derivative of the audio signal frame according to the frequency domain energy distribution of the audio signal frame; using the audio signal frame and each frame in a preset neighborhood range of the audio signal frame as a frame set, where the frame set includes a to-be-detected frame; and detecting the to-be-detected frame according to a maximum value distribution characteristic of a frequency domain energy distribution derivative of the frame set. In the various embodiments, detection on an audio signal can be implemented.