Method and apparatus for improving effective signal to noise ratios in hearing aids and other communication systems used in noisy environments without loss of spectral information

A signal enhancement apparatus for use in communication systems to enhance the effective signal to noise ratio of speech in a noisy environment without loss of spectral information and without introducing distortion into said speech, wherein amplitudes of said speech below a predetermined threshold are considered to be noise while amplitudes above the threshold are considered to be desired speech, said apparatus comprising:

• input means for providing an input electrical speech signal waveform having mixed speech and noise components temporally spaced by noise alone;
• an analog-to-digital converter for digitizing said input speech signal waveform;
• a memory for storing desired gain characteristics to attenuate portions of said input speech signal waveform having amplitudes below said predetermined threshold while passing portions of said speech signal waveform having amplitudes above the threshold substantially unchanged wherein the portions of the speech signal waveform below the threshold are attenuated such that the greater the input speech signal waveform amplitude is below the threshold, the greater the input speech signal waveform is attenuated;
• processing means for processing said digitized speech signal waveform in accordance with the desired gain characteristics to determine a digital control signal for controlling application of a gain to said input speech signal waveform;
• a digital-to-analog converter for converting said digital control signal to analog in order to control said application of said gain;
• and
• a two-quadrant multiplier for applying gain to said input speech signal waveform based on said analog control signal in order to reduce said noise in said speech signal waveform.

A signal enhancement apparatus for use in communication systems to enhance the effective signal to noise ratio of speech in a noisy environment without loss of spectral information and without introducing distortion into said speech, wherein amplitudes of said speech below a predetermined threshold are considered to be noise while amplitudes above the threshold are considered to be desired speech, said apparatus comprising:

• input means for providing an input electrical speech signal waveform having mixed speech and noise components temporally spaced by noise alone;
• an analog-to-digital converter for digitizing said input speech signal waveform;
• a memory for storing a desired gain response to attenuate portions of said input speech signal waveform having amplitudes below said predetermined threshold while passing portions of said speech signal waveform having amplitudes above the threshold substantially unchanged wherein the portions of the speech signal waveform below the threshold are attenuated such that the greater the input speech signal waveform amplitude is below the threshold, the greater the input speech signal waveform is attenuated;
• processing means for processing said digitized speech signal waveform in accordance with the desired gain response to reduce said noise in said input speech signal waveform;
• and
• a digital to analog converter for converting said digital reduced noise speech signal waveform from said processing means to analog in order to produce an acoustic output signal.

In a signal enhancement apparatus for use in communication systems, a method to enhance the effective signal to noise ratio of speech in a noisy environment without loss of spectral information and without introducing distortion into said speech, wherein amplitudes of said input speech signal waveform below a predetermined threshold are considered to be noise while amplitudes above the threshold are considered to be desired speech, said method comprising the steps of:

• (a) providing an input electrical speech signal waveform having mixed speech and noise components temporally spaced by noise alone;
• (b) digitizing said input speech signal waveform;
• (c) storing desired gain characteristics in a memory to attenuate portions of said input speech signal waveform having amplitudes below said predetermined threshold while passing portions of said speech signal waveform having amplitudes above the threshold substantially unchanged wherein the portions of the speech signal waveform below the threshold are attenuated such that the greater the input speech signal waveform amplitude is below the threshold, the greater the input speech signal waveform is attenuated;
• (d) processing said digitized speech signal waveform in accordance with the desired gain characteristics to determine a digital control signal for controlling application of a gain to said input speech signal waveform;
• (e) converting said digital control signal to analog in order to control said application of said gain;
• and
• (f) applying gain to said input speech signal waveform based on said analog control signal in order to reduce said noise in said speech signal waveform.

In a signal enhancement apparatus for use in communication svstems, a method to enhance the effective signal to noise ratio of speech in a noisy environment without loss of spectral information and without introducing distortion into said speech, wherein amplitudes of said input speech signal waveform below a predetermined threshold are considered to be noise while amplitudes above the threshold are considered to be desired speech, said method comprising the steps of:

• (a) providing an input electrical speech signal waveform having mixed speech and noise components temporally spaced by noise alone;
• (b) digitizing said input speech signal waveform;
• (c) storing a desired gain response in a memory to attenuate portions of said input speech signal waveform having amplitudes below said predetermined threshold while passing portions of said speech signal waveform having amplitudes above the threshold substantially unchanged wherein the portions of the speech signal waveform below the threshold are attenuated such that the greater the input speech signal waveform amplitude is below the threshold, the greater the input speech signal waveform is attenuated;
• (d) processing said digitized speech signal waveform in accordance with the desired gain characteristics to reduce said noise in said input speech signal waveform;
• and
• (e) converting said digital reduced noise speech signal waveform to analog in order to produce an acoustic output sianal.