A method comprising: receiving a first microphone signal from a first microphone having a first frequency response characteristic (1101, 1121) at frequencies (114) associated with wind noise; receiving a second microphone signal from a second microphone having a second frequency response characteristic (1102, 1122) at frequencies (114) associated with wind noise, wherein the first frequency response characteristic (1101, 1121) provides less gain than the second frequency response characteristic (1102, 1122) over the range of frequencies (114) associated with wind noise; and processing the first microphone signal and the second microphone signal to detect the presence of wind noise.