ECHO AND NOISE CANCELLATION

• US 20070274535A1
• Filed: 05/04/2006
• Published: 11/29/2007
• Est. Priority Date: 05/04/2006
• Status: Active Grant

• Current Assignee: Sony Interactive Entertainment Inc. (Sony Group Corp.)
• Original Assignee: Sony Computer Entertainment Incorporated (Sony Group Corp.)
• Inventors: Mao, Xiadong
• Granted Patent: US 7,545,926 B2
• Time in Patent Office: Days
• US Class Current: 381/94.100
• CPC Class Codes: H04M 9/082 using echo cancellers echo cancellers per se H04B3/23

Abstract

Methods and apparatus for echo cancellation in a system having a speaker and a microphone are disclosed. The speaker receives a speaker signal x(t). The microphone receives a microphone signal d(t) containing a local signal s(t) and an echo signal x₁(t) that is dependent on the speaker signal x(t). The microphone signal d(t) is filtered in parallel with first and second adaptive filters having complementary echo cancellation properties relative to each other. A minimum echo output e₃(t) is determined from an output e₁(t) of the first adaptive filter and an output e₂(t) of the second adaptive filter. The minimum echo output has a smaller energy and less correlation to the speaker signal x(t). A microphone output is then generated using the minimum echo output e₃(t).

Claims

1. A method for echo cancellation in a system having a speaker and a microphone, the speaker receiving a speaker signal x(t), the microphone receiving a microphone signal d(t) containing a local signal s(t) and an echo signal x₁(t), wherein the echo signal x₁(t) is dependent on the speaker signal x(t), the method comprising:

filtering the microphone signal d(t) in parallel with a first adaptive filter and a second adaptive filter, wherein the first adaptive filter has complementary echo cancellation properties relative to the second adaptive filter;

determining a minimum echo output e₃(t) from an output e₁(t) of the first adaptive filter and an output e₂(t) of the second adaptive filter, wherein the minimum echo output has a smaller energy and less correlation to the speaker signal x(t); and

generating a microphone output using the minimum echo output e₃(t).

2. The method of claim 1 wherein filtering the microphone signal d(t) in parallel with the first adaptive filter and the second adaptive filter includes adapting a set of filter coefficients of the first adaptive filter when a magnitude of x(t) is above a threshold and performing cross-correlation analysis between d(t) and x(t) with the second adaptive filter.

3. The method of claim 2 wherein determining a minimum echo output e₃(t) includes determining whether the second adaptive filter is excessively filtering out the local signal s(t);

and if the second adaptive filter is excessively filtering out the local signal s(t) using the output of the first adaptive filter as the minimum echo output.

4. The method of claim 3 wherein determining whether the second adaptive filter is excessively filtering out the local signal s(t) includes performing a cross-correlation between the output of the second adaptive filter and the speaker signal x(t).

5. The method of claim 3 further comprising, if the second adaptive filter is excessively filtering out the local signal s(t), replacing a set of filter coefficients for the second adaptive filter with a set of filter coefficients for the first echo cancellation filter and reinitializing the second echo cancellation filter to zero or a previous known state for which the second echo cancellation filter is not excessively filtering out the local signal s(t).

6. The method of claim 1 wherein generating a microphone output using the minimum echo output e₃(t) includes determining in parallel a first residual echo estimation ER₁(t) that includes a cross-correlation analysis between the minimum echo output e₃(t) and the speaker signal x(t) and determining a second residual echo estimation ER₂(t) that includes an echo-distance mismatch between the minimum echo output e₃(t) and the speaker signal x(t).

7. The method of claim 6 wherein the cross-correlation analysis includes a computation of an expectation value of a cross-correlation between e₃(t) and x(t).

8. The method of claim 6 wherein the echo distance mismatch includes a computation of an expectation value of {e₃²(t)/x²(t)}.

9. The method of claim 6 wherein determining a first residual echo estimation ER₁(t) includes computing a second order norm N(1) of a first residual echo estimator ER(1) and wherein determining a second residual echo estimation ER₂(t) includes computing a second order norm N(2) of a second residual echo estimator ER(2), the method further comprising, during a period of double-talk, applying to e₃(t) the echo residual estimator ER(1) or ER(2) for which the corresponding norm N(1) or N(2) is smaller.

10. The method of claim 6 wherein determining a first residual echo estimation ER₁(t) includes computing a second order norm N(1) of a first residual echo estimator ER(1) and wherein determining a second residual echo estimation ER₂(t) includes computing a second order norm N(2) of a second residual echo estimator ER(2), the method further comprising, during a period of single-talk, applying to e₃(t) the echo residual estimator ER(1) or ER(2) for which the corresponding norm N(1) or N(2) is larger.

11. The method of claim 6 wherein generating a microphone output using the minimum echo output e₃(t) further includes determining a minimum residual echo estimation ER₃(t), wherein the minimum residual echo estimation ER₃(t) is the one of ER₁(t) and ER₂(t) having a minimum energy and minimum correlation to x(t).

12. The method of claim 11 further comprising and selectively removing ER₃(t) from the minimum echo output e₃(t).

13. The method of claim 12 wherein selectively removing ER₃(t) from the microphone output includes maximally removing ER₃(t) from the microphone output during a far-end-talk only period.

14. The method of claim 12 wherein selectively removing ER₃(t) from the microphone output includes minimally removing ER₃(t) from the microphone output if the local signal s(t) has a non-zero magnitude.

15. The method of claim 1 further comprising computing an estimated noise signal n′

(t) from the microphone signal d(t); and

reducing a level of the estimated noise signal n′

(t) to form a reduced noise signal n″

(t); and

incorporating the reduced noise signal n″

(t) into the microphone output signal.

16. An echo cancellation apparatus for use in a system having a speaker and a microphone, the speaker being adapted to receive a speaker signal x(t), the microphone being adapted to generate a microphone signal d(t) containing a local signal s(t) and an echo signal x₁(t), wherein the echo signal x₁(t) is dependent on the speaker signal x(t), the apparatus comprising:

a first adaptive filter coupled to the speaker and the microphone; and

a second adaptive filter coupled to the speaker and the microphone in parallel with the first adaptive filter, wherein the second adaptive filter has complementary echo cancellation properties relative to the first adaptive filter;

an integrator coupled to the first adaptive filter and the second adaptive filter, wherein the integrator is configured to determine a minimum echo output e₃(t) from an output e₁(t) of the first adaptive filter and an output e₂(t) of the second adaptive filter, wherein the minimum echo output has a smaller energy and less correlation to the speaker signal x(t).

17. The apparatus of claim 16 wherein the integrator is configured to determine whether the second echo cancellation filter is excessively filtering out the local signal s(t);

wherein if the second adaptive filter is excessively filtering out the local signal s(t) integrator selects the e₁(t) as the minimum echo output e₃(t).

18. The apparatus of claim 16 wherein the first adaptive filter is a voice activity detector filter.

19. The apparatus of claim 16 wherein the second adaptive filter is a cross-correlation analysis filter.

20. The apparatus of claim 16, further comprising a first echo residual estimator ER(1) coupled to the integrator and a second echo residual estimator ER(2) coupled to the integrator.

21. The apparatus of claim 20 wherein the first echo residual estimator ER(1) is configured to generate a first residual echo estimation ER₁(t) that includes a cross-correlation analysis between the minimum echo output e₃(t) and the speaker signal x(t) and wherein the second echo residual estimator ER(2) is configured to determine a second residual echo estimation ER₂(t) that includes an echo-distance mismatch between the minimum echo output e₃(t) and the speaker signal x(t).

22. The apparatus of claim 21 wherein the cross-correlation analysis includes a computation of an expectation value of a cross-correlation between e₃(t) and x(t).

23. The apparatus of claim 21 wherein the echo distance mismatch includes a computation of an expectation value of {e₃²(t)/x²(t)}.

24. The apparatus of claim 21, further comprising a residual echo cancellation module coupled to the first and second echo residual estimators, the residual echo cancellation module is configured to compute a second order norm N(1) of the first residual echo estimator ER(1) and compute a second order norm N(2) of the second residual echo estimator ER(2), the echo cancellation module be configured to apply to e₃(t), during a period of double-talk, the echo residual estimator ER(1) or ER(2) for which the corresponding norm N(1) or N(2) is smaller.

25. The apparatus of claim 21, further comprising a residual echo cancellation module coupled to the first and second echo residual estimators, the residual echo cancellation module is configured to compute a second order norm N(1) of the first residual echo estimator ER(1) and compute a second order norm N(2) of the second residual echo estimator ER(2), the echo cancellation module be configured to apply to e₃(t), during a period of single-talk, the echo residual estimator ER(1) or ER(2) for which the corresponding norm N(1) or N(2) is larger.

26. The apparatus of claim 21, further comprising a residual echo cancellation module coupled to the first and second echo residual estimators, the residual echo cancellation module is configured to determine a minimum residual echo estimation ER₃(t), wherein the minimum residual echo estimation ER₃(t) is the one of ER₁(t) and ER₂(t) having a minimum energy and minimum correlation to x(t).

27. The apparatus of claim 26 wherein the residual echo cancellation module is configured to selectively remove ER₃(t) from the minimum echo output e₃(t).

28. The apparatus of claim 27 wherein the residual echo cancellation module is configured to maximally removing ER₃(t) from the microphone output during a far-end-talk only period.

29. The apparatus of claim 27 wherein the residual echo cancellation module is configured to minimally removing ER₃(t) from the microphone output if the local signal s(t) has a non-zero magnitude.

30. The apparatus of claim 16 further comprising a noise cancellation module coupled to the microphone, wherein the noise cancellation module is configured to compute an estimated noise signal n′

(t) from the microphone signal d(t);

reduce a level of the estimated noise signal n′

(t) to form a reduced noise signal n″

(t); and

incorporate the reduced noise signal n″

(t) into a microphone output signal.

31. An audio signal processing system, comprising:

a microphone;

a speaker;

a processor coupled to the microphone and the speaker;

a memory coupled to the processor, the memory having embodied therein a set of processor readable instructions for implementing a method for method for echo cancellation in a system having a speaker and a microphone, the speaker receiving a speaker signal x(t), the microphone receiving a microphone signal d(t) containing a local signal s(t) and an echo signal x₁(t), wherein the echo signal x₁(t) is dependent on the speaker signal x(t), the processor readable instructions including;

instructions for filtering the microphone signal d(t) in parallel with a adaptive filter and a second adaptive filter, wherein the first adaptive filter has complementary echo cancellation properties relative to the second adaptive filter;

instructions for determining a minimum echo output e₃(t) from an output e₁(t) of the first adaptive filter and an output e₂(t) of the second adaptive filter, wherein the minimum echo output has a smaller energy and less correlation to the speaker signal x(t); and

instructions for generating a microphone output using the minimum echo output e₃(t).

32. A processor readable medium having embodied therein a memory coupled to the processor, the memory having embodied therein a set of processor readable instructions for implementing a method for method for echo cancellation in a system having a speaker and a microphone, the speaker receiving a speaker signal x(t), the microphone receiving a microphone signal d(t) containing a local signal s(t) and an echo signal x₁(t), wherein the echo signal x₁(t) is dependent on the speaker signal x(t), the processor readable instructions including:

instructions for filtering the microphone signal d(t) in parallel with a adaptive filter and a second adaptive filter, wherein the first adaptive filter has complementary echo cancellation properties relative to the second adaptive filter;

instructions for determining a minimum echo output e₃(t) from an output e₁(t) of the first adaptive filter and an output e₂(t) of the second adaptive filter, wherein the minimum echo output has a smaller energy and less correlation to the speaker signal x(t); and

instructions for generating a microphone output using the minimum echo output e₃(t).