Speech or sound recognition device

摘要

Sound is converted into an electrical speech signal which is fed to a processor to transform the signal in a predetermined way so as to produce a speech model in the form of a time-frequency distribution. A detector determines a section of the model in which the speech energy of the sound is equal to or less than a first value, and in which the low frequency component of the sound is equal to or greater than a second value. The speech model is superimposed over the already produced speech model for the same sound according to the conclusion of the section. Pref. the section chosen is the hiss-component of the sound, and the converter device is a microphone.