We have developed a technique by which devices like Pocket PCs, phones, or laptop computers can use audio processing to estimate their locations relative to one another. This will enable the creation of ad hoc speaker or microphone arrays from heterogeneous collections of mobile devices.