 Feature detectors are specialized receptors for specific properties of the incoming signal, for example for F2 transition or voice onset time. They serve as the basis for the recognition product. And this is how perception would work. First, the ear amplifies the incoming signal and transmits it to the auditory nerve. During the primary auditory analysis, the signal is filtered and non-speech components are separated from speech. To cope with a complex signal, auditory patterns such as formant frequencies must be identified. At the end of the recognition process, phonological units such as phonemes or features are extracted from the signal. The whole process is coordinated by the brain, which uses a set of feature detectors to process the incoming signal.