Auto-tuning of an acoustic echo canceller

Patent metadata

Field	Value
Publication number	US-9692882-B2
Application number	US-201514677693-A
Country	US
Kind code	B2
Filing date	Apr 2, 2015
Priority date	Apr 2, 2014
Publication date	Jun 27, 2017
Grant date	Jun 27, 2017

Abstract

A gain control system for dynamically tuning an echo canceller, the echo canceller being configured to estimate an echo of a far-end signal and subtract that estimate from a microphone signal to output an echo cancelled signal, the system comprising an echo measurement unit configured to calculate a ratio of the microphone signal to the far-end signal, an attenuation unit configured to attenuate at least one of the microphone signal and the far-end signal to output a second microphone signal and a second far-end signal to the echo canceller, the ratio of the second microphone signal to the second far-end signal being different from the calculated ratio, and an attenuation controller configured to control the attenuation unit, in dependence on the calculated ratio, so as to alter the ratio of the second microphone signal to the second far-end signal and control the echo-cancelled signal.

First claim

The invention claimed is:

1. A gain control system for dynamically tuning an echo canceller, the echo canceller being configured to estimate an echo of a far-end signal and subtract that estimate from a microphone signal to output an echo cancelled signal, the system comprising:
an echo measurement unit configured to calculate a ratio of a first microphone signal to a first far-end signal;
an attenuation unit configured to attenuate at least one of the first microphone signal and the first far-end signal to output a second microphone signal and a second far-end signal to the echo canceller, the ratio of the second microphone signal to the second far-end signal being different from the calculated ratio; and
an attenuation controller configured to control the attenuation unit, in dependence on the calculated ratio, so as to alter the ratio of the second microphone signal to the second far-end signal and control the echo-cancelled signal.

2. A gain control system as claimed in claim 1 , in which the echo canceller has multiple different modes of operation, the attenuation controller being configured to associate each mode of operation of the gain control system with default attenuations of the first microphone signal and the first far-end signal.

3. A gain control system as claimed in claim 2 , the attenuation controller being configured to:
at the commencement of an estimation, control the attenuation unit to apply the default attenuations associated with the echo canceller's mode of operation; and
thereafter, control the attenuation unit in dependence on the calculated ratio.

4. A gain control system as claimed in claim 1 , the attenuation controller being configured to determine that an estimation has commenced if it determines that one or more of the following has occurred: a change in the mode of operation of the echo canceller; the start of a communication by the communication device; a change in the volume level of the communication device.

5. A gain control system as claimed in claim 1 , comprising a presence unit configured to, before the echo measurement unit calculates the ratio, check whether a signal energy of the first far-end signal is above a predetermined threshold, the echo measurement unit being configured to, if the signal energy of the first far-end signal is not above the predetermined threshold, not calculate the ratio until the signal energy of the first far-end signal is found to be above the predetermined threshold.

6. A gain control system as claimed in claim 1 , the echo canceller comprising an adaptive filter configured to generate an estimate of the echo from the second far-end signal, and the gain control system comprising an adaptation counter configured to maintain a count of adaptation by the adaptive filter;
the adaptation counter being configured to, before the echo measurement unit calculates the ratio, check whether the count is above a predetermined threshold; and
the echo measurement unit being configured to, if the count is not above the predetermined threshold, not calculate the ratio until the count is found to be above the predetermined threshold.

7. A gain control system as claimed in claim 1 , the gain control system comprising a mode controller configured to, if the count is not above the predetermined threshold, control the echo canceller to operate in half-duplex mode.

8. A gain control system as claimed in claim 1 , the gain control system comprising a single talk discriminator configured to determine whether the first microphone signal is indicative of single talk; the single talk discriminator being configured to, before the echo measurement unit calculates the ratio, check whether the first microphone signal is indicative of single talk; and the echo measurement unit being configured to, if the first microphone signal is determined not to be indicative of single talk, not calculate the ratio until the first microphone signal is determined to be indicative of single talk.

9. A gain control system as claimed in claim 8 , the gain control system comprising a mode controller configured to, if the first microphone signal is determined not to be indicative of single talk, control the echo canceller to operate in full duplex mode if the echo canceller is already operating in that mode and to operate in half-duplex mode otherwise.

10. A gain control system as claimed in claim 8 , the single talk discriminator being configured to:
compute a first ratio of a short term power of the error signal to a short term power of the first microphone signal;
compare the first ratio with a first predetermined threshold; and
determine that the first microphone signal is indicative of single talk if the first ratio is above the first predetermined threshold.

11. A gain control system as claimed in claim 10 , the single talk discriminator being configured to, if the first ratio is not above the predetermined threshold:
compute a second ratio of a short term power of the error signal to a long term power of the first far-end signal;
compare the second ratio with a predetermined threshold; and
determine that the first microphone signal is indicative of single talk if the second ratio is above the second predetermined threshold.

12. A gain control system as claimed in claim 11 , the single talk discriminator being configured to determine that the first microphone signal is not indicative of single talk if the second ratio is not above the second predetermined threshold.

13. A gain control system as claimed in claim 1 , the echo measurement unit being configured to calculated the ratio of the first microphone signal to the first far-end signal by:
estimating a long-term average power of the first microphone signal and a long-term average power of the first far-end signal; and
calculating the ratio of the long-term average power of the first microphone signal to the long-term average power of the first far-end signal.

14. A gain control system as claimed in claim 1 , the gain control system comprising a confirmation unit configured to monitor the calculated ratio and halt the calculation if the calculated ratio is determined to have been stable for a predetermined length of time.

15. A gain control system as claimed in claim 14 , the gain control system comprising a mode controller configured to, if the calculated ratio is determined to have been stable for a predetermined length of time, control the echo canceller to operate in full duplex mode.

16. A gain control system as claimed in claim 1 , the attenuation controller being configured to control the attenuation unit in dependence on a control signal from a convergence unit that is configured to monitor the stability of the adaptive filter.

17. A method for dynamically tuning an echo canceller, the echo canceller being configured to estimate an echo of a far-end signal and subtract that estimate from a microphone signal to output an echo cancelled signal, the method comprising:
calculating a ratio of a first microphone signal to a first far-end signal;
attenuating at least one of the first microphone signal and the first far-end signal to output a second microphone signal and a second far-end signal to the echo canceller, the ratio of the second microphone signal to the second far-end signal thereby being different from the calculated ratio; and
controlling the attenuation unit, in dependence on the calculated ratio, so as to alter the ratio of the second microphone signal to the second far-end signal and control the echo-cancelled signal.

18. A method as claimed in clai

Assignees

• Imagination Tech Ltd

Inventors

• Mani Senthil Kumar
• Namani Gandhi
• Akella Srinivas
• Pulugurtha Sai Ravi Teja

Classifications

• H04M9/082 Primary

  using echo cancellers (echo cancellers per se H04B3/23) · CPC title

• H04M3/002 Primary

  Applications of echo suppressors or cancellers in telephonic connections (in two-way loud-speaking telephone systems H04M9/08, echo suppressors or cancellers per se H04B3/20) · CPC title

• H04M1/20

  Arrangements for preventing acoustic feed-back (H04M1/62 takes precedence) · CPC title

• H04B3/23

  using a replica of transmitted signal in the time domain, e.g. echo cancellers · CPC title

• H04M9/08 Primary

  Two-way loud-speaking telephone systems with means for conditioning the signal, e.g. for suppressing echoes for one or both directions of traffic · CPC title