Speech signal amplitude equalizer
AssigneeBell Telephone Laboratories, Incorporated
InventorsJames Loton Flanagan, Susan Anne Webber
Life of this patent
Estimating the term…
Abstract
A speech signal amplitude equalizer (compressor) circuit normalizes the amplitude variations of a real-time input speech signal s(t) to form the normalized output signal r(t) which may be either of two values: a constant amplitude (unilevel) speech signal or a zero-amplitude (squelched) signal. The normalized output signal r(t)=s(t)/a(t)+K where the envelope signal a(t)=s.sup.2 (t)+s.sup.2 (t) in which s.sup.2 (t) is the Hilbert transform and K is a control signal which can take one of two levels depending upon the amplitude of the envelope signal being above or below certain thresholds in a hysteresis type logic circuit.
Claims20
A signal conditioning circuit comprising a speech signal source;
- means responsive to a speech signal from said source for generating a signal representative of the envelope of said speech signal;
- a first threshold source;
- a second threshold source, said second threshold being less than said first threshold;
- means for producing a level signal of first value responsive solely to said envelope representative signal exceeding said first threshold and remaining above said second threshold and of second value responsive solely to said envelope representative signal decreasing below said second threshold and remaining below said first threshold;
- means for combining said envelope representative signal and said level signal to form a modified envelope signal;
- and means for dividing said speech signal by said modified envelope signal to produce a substantially unilevel amplitude output signal when said first value level signal is generated and a substantially zero valued output signal when said second value level signal is generated.
An amplitude equalizing circuit comprising means for receiving an input speech-like signal;
- means responsive to said input speech-like signal for generating an analytic signal of which said speech-like signal constitutes the real part, means responsive to said analytic signal for forming a first signal representative of the envelope of said speech-like signal;
- a first threshold source, a second threshold source, said first threshold being greater than said second threshold, means responsive solely to said first signal increasing above said first threshold and remaining above said second threshold for producing a first level signal;
- means responsive solely to said first signal decreasing below said second threshold and remaining below said first threshold for producing a second level signal;
- means for summing said first signal and the produced one of said first and second level signals to form a modified first signal;
- and means for forming an output signal corresponding to the ratio of the received input speech signal to said modified first signal, said output signal being a substantially parallel unilevel amplitude signal when said first level signal is produced and being a substantially zero valued signal when said second level signal is produced.
A circuit for transforming a speech signal into unilevel amplitude speech signal comprising means for receiving an input speech signal;
- means responsive to said received input speech signal for forming an analytic signal corresponding to said speech signal, means responsive to said analytic signal for generating a signal representative of the envelope of said speech signal;
- a first threshold level generator;
- a second threshold level generator, said second threshold level being less than said first threshold level;
- means solely responsive to said envelope representative signal exceeding said first threshold level and remaining greater than said second threshold level for producing a first constant value signal;
- means responsive to said envelope representative signal decreasing below said second threshold level and remaining less than said first threshold level for producing a second constant value signal;
- means for summing said envelope representative signal and the produced one of said first and second constant value signals to form a third signal;
- and means for dividing said received speech signal by said third signal to form a substantially unilevel amplitude speech signal when said first level signal is produced and a substantially zero level signal when said second level signal is produced.
A method for transforming a speech signal into a substantially unilevel amplitude speech signal comprising the steps of receiving an input speech signal, forming a signal representative of the envelope of said input speech signal, generating first and second thresholds, said first threshold being greater than said second threshold, producing a first level signal responsive solely to said envelope representative signal exceeding said first threshold and remaining above said second threshold, producing a second level signal responsive solely to said envelope representative signal decreasing below said second threshold and remaining below said first threshold, combining said envelope signal with the produced one of said first and second level signals, and dividing said input speech signal by said combined signal to produce a substantially unilevel amplitude signal when said first level signal is produced and a substantially zero amplitude signal when said second level signal is produced.
Dependencies are inferred from the claim text.