Speech processing system and method for enhancing a speech signal in a noisy environment
First Claim
1. A method for enhancing speech signals in a noisy environment, comprising the steps of:
- inputting a digital speech signal x(k) at an input of a plurality of successive delay elements whose outputs form a like plurality of taps of an adaptive finite impulse response (FIR) filter;
inputting said digital speech signal and each of said plurality of taps to corresponding inputs of a plurality of variable multipliers;
computing a first signal power estimate y(k) at a sample point k given by the formula y(k)=β
1 y(k-1)+(1-β
1)x2 (k);
computing a second signal power estimate z(k) at said sample point k given by the formula z(k)=β
2 z(k-1)+(1-β
2)x2 (k);
choosing a value of β
1 greater than a value of β
2 ;
selecting an overall signal power estimate yz(k) at said sample point k as a maximum one of said first signal power estimate y(k) and said second signal power estimate z(k);
recursively updating a plurality of FIR filter coefficients corresponding to said plurality of variable multipliers according to a normalized least-mean-squares (NLMS) prediction using said overall signal power estimate yz(k) and an estimation error signal to provide updated values of said plurality of FIR filter coefficients;
providing said updated values of said plurality of FIR filter coefficients to coefficient inputs of said plurality of variable multipliers; and
summing outputs of said plurality of variable multipliers to provide an enhanced speech signal as an output of said adaptive FIR filter.
18 Assignments
0 Petitions
Accused Products
Abstract
A speech processing system (30) operates in a noisy environment (20) by performing adaptive prediction between inputs from two sensors positioned to transduce speech from a speaker, such as an accelerometer and a microphone. An adaptive filter (37) such as a finite impulse response (FIR) filter receives a digital accelerometer input signal, adjusts filter coefficients according to an estimation error signal, and provides an enhanced speech signal as an output. The estimation error signal is a difference between a digital microphone input signal and the enhanced speech signal. In one embodiment, the adaptive filter (37) selects a maximum one of a first predicted speech signal based on a relatively-large smoothing parameter and a second predicted speech signal based on a relatively-small smoothing parameter, with which to normalize a predicted signal power. The predicted signal power is then used to adapt the filter coefficients.
148 Citations
9 Claims
-
1. A method for enhancing speech signals in a noisy environment, comprising the steps of:
-
inputting a digital speech signal x(k) at an input of a plurality of successive delay elements whose outputs form a like plurality of taps of an adaptive finite impulse response (FIR) filter; inputting said digital speech signal and each of said plurality of taps to corresponding inputs of a plurality of variable multipliers; computing a first signal power estimate y(k) at a sample point k given by the formula y(k)=β
1 y(k-1)+(1-β
1)x2 (k);computing a second signal power estimate z(k) at said sample point k given by the formula z(k)=β
2 z(k-1)+(1-β
2)x2 (k);choosing a value of β
1 greater than a value of β
2 ;selecting an overall signal power estimate yz(k) at said sample point k as a maximum one of said first signal power estimate y(k) and said second signal power estimate z(k); recursively updating a plurality of FIR filter coefficients corresponding to said plurality of variable multipliers according to a normalized least-mean-squares (NLMS) prediction using said overall signal power estimate yz(k) and an estimation error signal to provide updated values of said plurality of FIR filter coefficients; providing said updated values of said plurality of FIR filter coefficients to coefficient inputs of said plurality of variable multipliers; and summing outputs of said plurality of variable multipliers to provide an enhanced speech signal as an output of said adaptive FIR filter. - View Dependent Claims (2, 3, 4, 5)
-
-
6. An adaptive finite impulse response filter, comprising:
-
a signal input for receiving an input signal; a feedback input for receiving an error signal; an output for providing a filtered output signal; a plurality of filter taps coupled in series with said input signal; a plurality of variable multipliers for multiplying said input signal and outputs of said plurality of filter taps by corresponding ones of a plurality of coefficients; a summing device for providing said filtered output signal as a sum of outputs of each of said plurality of variable multipliers; and prediction means for receiving said error signal, and for recursively updating each coefficient according to a maximum one of first and second predicted values of said filtered output signal, in response to said error signal; said first predicted value of said filtered output signal having a first smoothing parameter associated therewith; said second predicted value of said filtered output signal having a second smoothing parameter associated therewith; said first smoothing parameter providing a faster prediction response after a change in said input signal than said second smoothing parameter. - View Dependent Claims (7)
-
-
8. An adaptive finite impulse response filter, comprising:
-
a signal input for receiving an input signal; a feedback input for receiving an error signal; an output for providing a filtered output signal; a plurality of filter taps coupled in series with said input signal; a plurality of variable multipliers for multiplying said input signal and outputs of each of said filter taps by corresponding ones of a plurality of coefficients; a summing device for providing said filtered output signal as a sum of outputs of each of said plurality of multipliers; and a predictor having an input for receiving said error signal, and outputs for providing corresponding ones of said plurality of coefficients; said predictor computing first and second predicted values of said output signal and recursively updating each of said plurality of coefficients as determined by a maximum one of said first and second predicted values of said output signal, and said error signal; said first predicted value characterized as providing a faster prediction response after a change in said input signal than said second predicted value, whereby the adaptive finite impulse response filter is characterized as having a fast attack and a slow release. - View Dependent Claims (9)
-
Specification