Microphone array processing for adaptive echo control

1. An apparatus comprising:

a beamformer configured to pass desired portions of audio signals received using two or more microphones and further configured to suppress undesired portions of the audio signals, wherein the beamformer comprises a speech blocking filter to prevent suppression of near-end desired talker speech in the audio signals and further comprises an echo suppression filter to suppress echo from the audio signals;

an echo suppression control unit coupled to the beamformer, wherein the echo suppression control unit is configured to receive a signal and determine whether to dynamically adapt the speech blocking filter or to dynamically adapt the echo suppression filter, wherein the speech blocking filter remains unchanged during dynamic adaptation of the echo suppression filter and wherein the echo suppression filter remains unchanged during dynamic adaptation of the speech blocking filter; and

a residual echo cancellation unit coupled to the beamformer, wherein the residual echo cancellation unit is configured to receive output audio signals from the beamformer and to further suppress residual echo in the output audio signals.

2. The apparatus as described by claim 1 further comprising a persistent interference detector configured to receive the audio signals and further configured to detect presence of persistence echo, and wherein the persistent interference detector is configured to generate a signal indicating whether a persistent echo is present, and wherein the signal is input to the echo suppression control unit that determines whether to dynamically adapt either the speech blocking filter or the echo suppression filter responsive to the signal.

3. The apparatus as described by claim 1 , wherein the signal is an output signal from the echo cancellation unit.

4. The apparatus as described by claim 1 further comprising a plurality of acoustic echo cancelers (AECs) coupled to an input of the beamformer, wherein the plurality of AECs is configured to reduce echo in the audio signals at the inputs of the beamformer by reducing power thereof.

5. The apparatus as described by claim 4 , wherein the echo suppression control unit is configured to maintain the echo suppression filter unchanged during startup of the apparatus, wherein the plurality of AECs adapt rapidly during the startup, and wherein the echo suppression control unit is configured to cause the beamformer to dynamically adapt the speech blocking filter during the startup of the apparatus if near speech is detected.

6. The apparatus as described by claim 1 , wherein the echo suppression control unit is configured to maintain the echo suppression filter unchanged in response to detecting a trigger-word.

7. The apparatus as described by claim 1 , wherein the echo suppression control unit is configured to cause the echo suppression filter to dynamically adapt in response to detecting that the received audio signals comprise an echo dominating portion and further responsive to detecting that playback is active.

8. The apparatus as described by claim 1 , wherein the echo suppression control unit is configured to cause the beamformer to dynamically adapt the speech blocking filter in response to detecting near speech and further in response to detecting presence of near speech dominating portion in the received audio signals.

9. An apparatus comprising:

a processor comprising a first filter and a second filter, wherein the processor is configured to receive audio signals from two or more microphones and wherein the processor is configured to suppress undesired portions of the audio signals and to pass desired portions of the audio signals; and

an echo suppression control unit configured to receive a signal and determine whether to dynamically adapt the first filter to prevent suppression of near-end desired talker speech in the received audio signals or the second filter to suppress echo in the received audio signals, wherein the second filter remains unchanged during dynamic adaptation of the first filter and wherein the first filter remains unchanged during dynamic adaptation of the second filter.

10. The apparatus as described by claim 9 , wherein the first filter of the processor is a speech blocking filter and wherein the second filter of the processor is an echo suppression filter.

11. The apparatus as described by claim 9 further comprising a plurality of acoustic echo cancelers (AECs) coupled to the input of the processor, wherein the plurality of AECs is configured to reduce echo in the audio signals at the inputs of the processor by reducing power thereof.

12. The apparatus as described by claim 11 , wherein the echo suppression control unit is configured to maintain the second filter of the processor unchanged during startup of the apparatus, wherein the plurality of AECs adapt rapidly during the startup, and wherein the echo suppression control unit is configured to cause the processor to dynamically adapt the first filter of the processor during the startup of the apparatus if near speech is detected.

13. The apparatus as described by claim 9 , wherein the echo suppression control unit is configured to maintain the second filter of the processor unchanged in response to detecting a trigger word.

14. The apparatus as described by claim 9 , wherein the echo suppression control unit is configured to cause the second filter of the processor to dynamically adapt in response to detecting that the received audio signals comprise an echo dominating portion and further responsive to detecting that a signal associated with playback is active.

15. The apparatus as described by claim 9 , wherein the echo suppression control unit is configured to cause the first filter of the processor to dynamically adapt in response to detecting near speech and further in response to detecting presence of near speech dominating portion in the received audio signals.

16. A method comprising:

receiving audio signals using two or more microphones;

determining whether near speech is present;

determining whether the audio signals comprise echo dominating portions;

responsive to determining presence of the near speech, dynamically adapting a speech blocking filter;

responsive to determining presence of the echo dominating portions, dynamically adapting an echo suppression filter, wherein the speech blocking filter remains unchanged during adaptation of the echo suppression filter and wherein the echo suppression filter remains unchanged during adaptation of the speech blocking filter;

passing desired portions of the audio signals; and

suppressing undesired portions of the audio signals.

17. The method as described by claim 16 further comprising further suppressing residual echo in the audio signals.

18. The method as described by claim 16 further comprising determining whether persistent echo is present and in response to determining that persistent echo is present dynamically adapting the echo suppression filter.

19. The method as described by claim 16 further comprising:

maintaining the echo suppression filter unchanged responsive to detecting rapid adaptation of acoustic echo cancelers during startup; and

responsive to determining presence of near speech, dynamically adapting the speech blocking filter.

20. The method as described by claim 16 further comprising maintaining the echo suppression filter unchanged in response to detecting a trigger word.

21. The method as described by claim 16 further comprising dynamically adapting the echo suppression filter in response to detecting that the received audio signals comprise an echo dominating portion and further responsive to detecting that playback is active.

22. The method as described by claim 16 further comprising dynamically adapting the speech blocking filter in response to detecting near speech and further in response to detecting presence of near speech dominating portion in the received audio signals.