Tone detector with improved performance in the presence of speech

摘要

This tone detector which is suited for dual tone multifrequency (DTMF) digit detection, separates the pairs of tones into a low frequency and high frequency group in which the other frequency group is attenuated by filtering. The amplitude of the low and high frequency groups serve as separate variable threshold reference signals. Individual tones are bandpass filtered with their corresponding outputs converted into an amplitude signal. The amplitude of the individually bandpass filtered tone signals are compared against the corresponding low or high frequency variable threshold signal for a determination of whether a tone is present. Preferably, additional validation steps are applied to the preliminary tone decisions to provide improved digit detection accuracy, especially in the presence of other signals such as speech. <IMAGE>