Identification of spoken words based on aggregated knowledge of the speech characteristics of a population of speakers. See also speaker-dependent speech recognition.