The present technology relates to an acoustic signal processing apparatus, an acoustic signal processing method and a program which can widen the variations of the configuration of a virtual surround system that stabilizes the localization sensation of a virtual speaker.Crosstalk correction processing is performed on a first binaural signal based on a sound source opposite side HRTF and a second binaural signal based on a sound source side HRTF. A first acoustic signal and a second acoustic signal are generated. A component of a first frequency band, in which a first notch of the sound source opposite side HRTF appears, and a component of a second frequency band, in which a second notch appears, are attenuated in an input signal or the second binaural signal, thereby attenuating the component of the first frequency band and the component of the second frequency band of the first acoustic signal and the second acoustic signal. An auxiliary signal including a component of a third frequency band of the input signal or the second binaural signal, in which the component of the first frequency band and the component of the second frequency band are attenuated, is added to the first acoustic signal, and a third acoustic signal is generated. The present technology can be applied to, for example, an AV amplifier.