Signal rate synchronization for remote acoustic echo cancellation

• US 9,373,318 B1
• Filed: 03/27/2014
• Issued: 06/21/2016
• Est. Priority Date: 03/27/2014
• Status: Expired due to Fees

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Narayanan, Preethi Parasseri, Beddingfield, Gabe, Piersol, Kurt Wesley, Ayrapetian, Robert, Charton, Arnaud Jean-Louis, Pogue, Michael Alan, Su, Yuwen
• Primary Examiner(s): Chin, Vivian
• Assistant Examiner(s): Kurr, Jason R
• Application Number:
  US14/228,045
• Time in Patent Office: 817 Days
• Field of Search: 381/66, 381/71.1, 381/71.8, 381/71.9, 381/71.11, 381/71.13, 381/94.1, 381/94.4, 381/94.7, 700/94, 455/501, 455/502, 379/406.01, 379/406.02, 379/406.06, 379/406.08, 379/406.1
• US Class Current: 1/1
• CPC Class Codes: G10K 11/175 using interference effects; Masking sound
• CPC Class Codes: G10L 2021/02082 the noise being echo, reverberation of the speech
• CPC Class Codes: G10L 2021/02166 Microphone arrays; Beamforming
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: G10L 21/0232 Processing in the frequency domain
• CPC Class Codes: G10L 21/0264 characterised by the type of parameter measurement, e.g. correlation techniques, zero crossing techniques or predictive techniques
• CPC Class Codes: H04R 3/02 for preventing acoustic reaction , i.e. acoustic oscillatory feedback specially adapted for hearing aids H04R25/453

Abstract

A system may be configured to interact with a user through speech using a first and second audio devices, where the first device produces audio and the second device captures audio. The second device may be configured to perform acoustic echo cancellation with respect to a microphone signal based on a reference signal provided by the first device. The reference and microphone signals may have the same nominal signal rates. However, the signal rates may drift from each other over time. In order to synchronize the rates of the signals, each of the devices maintains a signal index. The second device compares the values of the two signal indexes over time to determine rate differences between the reference and microphone signals and then corrects for the rate differences.

Claims

1. A method, comprising:

producing output audio at a speaker of a first device;

generating reference clock cycles at the first device to establish a first signal rate;

generating a digital reference signal that contains the output audio at the first signal rate;

maintaining a reference index at the first device, the reference index having a value that is incremented in response to the reference clock cycles;

providing the digital reference signal to a second device;

repeatedly providing a current value of the reference index to the second device;

generating input clock cycles at the second device to establish a second signal rate;

generating a digital audio signal from a microphone of the second device at the second signal rate, wherein the digital audio signal contains the output audio produced by the first device;

maintaining an input index at the second device, the input index having a value that is incremented in response to the input clock cycles;

determining a rate of change of the reference index and a rate of change of the input index based at least in part on the repeatedly provided current value of the reference index and a corresponding current value of the input index;

comparing the rate of change of the reference index and the rate of change of the input index to determine a signal rate difference between the first and second signal rates;

increasing or decreasing at least one of the second signal rate of the digital audio signal or the first signal rate of the digital reference signal to reduce the signal rate difference; and

performing acoustic echo cancellation at the second device in response to the digital reference signal to suppress the output audio in the digital audio signal.

2. The method of claim 1, wherein:

each of the digital audio signal and the digital reference signal comprises a series of signal values;

increasing the at least one of the second signal rate of the digital audio signal or the first signal rate of the digital reference signal comprises repeatedly adding signal values to at least one of the digital audio signal and the digital reference signal; and

decreasing the at least one of the second signal rate of the digital audio signal or the first signal rate of the digital reference signal comprises repeatedly removing signal values from at least one of the digital audio signal and the digital reference signal.

3. The method of claim 2, wherein the added signal values comprise duplicated signal values.

4. The method of claim 2, wherein the added signal values comprise interpolated signal values.

5. A first device, comprising:

a microphone that produces an analog audio signal containing first audio from a speaker of a second device;

a conversion component that converts the analog audio signal to a digital audio signal having a first signal rate;

a counter that produces an input index having a value that changes in synchronization with the first signal rate;

an acoustic echo canceller configured to receive a digital reference signal from the second device, the digital reference signal containing the first audio and having a second signal rate, wherein the acoustic echo canceller is responsive to the digital reference signal to suppress the first audio in the digital audio signal; and

one or more correction components configured to;

receive first and second values of a reference index from the second device at first and second times, respectively, wherein the values of the reference index change in synchronization with the second signal rate;

compare the first and second values of the reference index to corresponding first and second values of the input index at the first and second times, respectively, to determine a signal rate difference between the digital reference signal and the digital audio signal; and

process at least one of the digital audio signal or the digital reference signal to reduce the signal rate difference.

6. The first device of claim 5, wherein the conversion component converts the analog audio signal to the digital audio signal in response to a clock signal, wherein the counter increments the value of the input index in response to cycles of the clock signal.

7. The first device of claim 5, wherein the one or more correction components compare a change of the first and second values of the reference index and a change of the corresponding first and second values of the input index to determine the signal rate difference.

8. The first device of claim 5, wherein the one or more correction components average differences between changes in the first and second values of the reference index and changes in the first and second values of the input index to determine the signal rate difference.

9. The first device of claim 5, wherein the one or more correction components perform a linear regression analysis of the first and second values of the reference index versus the first and second values of the input index to determine the signal rate difference.

10. The first device of claim 5, wherein:

the digital reference signal is received in groups of reference signal values; and

the first and second values of the reference index are associated with first and second groups of the reference signal values, respectively.

11. The first device of claim 5, wherein the one or more correction components are further configured to process said at least one of the digital audio signal or the digital reference signal by removing values from at least one of the digital audio signal or the digital reference signal.

12. A method, comprising:

obtaining an analog signal via a microphone at a first device, the analog signal representing audio output by a speaker of a second device;

obtaining a first digital signal having a first signal rate, the first digital signal based at least in part on the analog signal;

producing a first index having values that change in synchronization with the first signal rate;

receiving a second digital signal having a second signal rate;

receiving first and second values of a second index, wherein the values of the second index change in synchronization with the second signal rate;

comparing the first and second values of the second index to corresponding first and second values of the first index to determine a rate difference between the first signal rate and the second signal rate;

processing at least one of the first and second digital signals to reduce the rate difference; and

performing acoustic echo cancellation at the first device to suppress at least a part of the audio in the first digital signal.

13. The method of claim 12, further comprising:

converting the analog signal to the first digital signal in response to a clock signal; and

wherein

producing the first index comprises counting cycles of the clock signal.

14. The method of claim 12, wherein the comparing comprises comparing a change of the first and second values of the second index and a change of the first and second values of the first index to determine the rate difference.

15. The method of claim 12, wherein the comparing comprises averaging differences between changes in the first and second values of the second index and changes in the first and second values of the first index to determine the rate difference.

16. The method of claim 12, wherein the comparing comprises performing a linear regression analysis of the first and second values of the second index versus the first and second values of the first index to determine the rate difference.

17. The method of claim 12, wherein:

the second digital signal is received in groups of signal values; and

each of the first and second values of the second index is associated with a corresponding group of signal values of the second digital signal.

18. The method of claim 12, wherein processing at least one of the first and second digital signals comprises removing values from at least one of the first digital signal or the second digital signal.

19. The method of claim 12, wherein performing the acoustic echo cancellation further comprises:

generating an echo-cancelled signal; and

providing the echo-cancelled signal to a remote computing device.

20. The method of claim 12, further comprising receiving the second digital signal from the second device.