Noise-discriminating voice-switched two-way intercom system

Claim 1

An intercommunication system for a banking installation or the like comprising in combination: customer speech transducing and sound reproduction means located at a customer location; a teller speech microphone and a speaker located at a teller location subjected to ambient noise, said teller speech microphone and speaker being separate electrical components, said teller speech microphone being physically positioned at a first spatial point at said teller location accessible to said teller for facilitating placement of the mouth of said teller, which constitutes a source of teller speech, proximate said teller speech microphone when said teller wishes to speak; power amplifier means with an input and an output for amplifying audio signals; a teller speech preamplifier means connected to said teller speech microphone; switch means interconnecting said teller speech preamplifier, said teller speaker and said customer speech transducing and sound reproduction means, said switch means being normally operative in a customer speech transmission mode to connect said customer speech transducing means to said power amplifier input and to connect said power amplifier output to said teller speaker to facilitate transmission of customer speech from said customer location to said teller location, said switch means being selectively operative in a teller speech transmission mode to connect said teller speech preamplifier to said power amplifier input and to connect said power amplifier output to said customer sound reproduction means to facilitate transmission of teller speech from said teller location to said customer location; switch control means including: a. a two state threshold detector having an output connected to said switch means, said threshold detector being normally operative in a customer state to place said switch means in said customer speech transmission mode and selectively operative in a teller state to place said switch means in said teller speech transmission mode, b. a noise microphone located at said teller location at a second spatial point spaced from said first spatial point whereat said teller speech microphone is located, c. a noise preamplifier connected to said noise microphone, d. noise amplitude extractor means connected to said noise preamplifier for providing an output signal having a level correlated to the amplitude of amplified noise output from said noise preamplifier, said output signal level being substantially independent of the frequency and phase of said amplified noise output from said noise preamplifier, and e. control signal means responsive to the outputs of said noise amplitude extractor means and said teller speech preamplifier for providing a control signal to said threshold detector when said outputs of said noise amplitude extractor means and said teller speech preamplifier bear a predetermined relation for switching said threshold detector to its teller state which in turn causes said switch means to switch to its teller speech transmission mode; said noise microphone, noise preamplifier and noise amplitude extractor means defining a noise channel, said teller speech microphone and teller speech preamplifier defining a teller speech channel; and means establishing the amplifier gain and microphone sensitivity of said noise and teller speech channels to cause the locus of points of different sound source positions, whereat equal outputs of said noise and teller speech channels are produced by a sound source, to form a closed surface which envelopes said teller speech microphone while excluding from the interior thereof said noise microphone, whereby a noise source lying exteriorly of said closed surface cannot cause said outputs of said noise amplitude extractor means and teller speech preamplifier to bear said predetermined relation and produce said control signal to switch said threshold detector to its teller state and in turn switch said switch means to its teller speech transmission mode, and whereby further only a source of teller speech located within said closed surface in the absence of a noise source within said closed surface can cause said outputs of said noise amplitude extractor means and teller speech preamplifier to bear said predetermined relation to cause said threshold detector to switch to its teller state and in turn said switch means to switch to its teller speech transmission mode to facilitate transmission of teller speech from said teller location to said customer location.