Method and apparatus for active reduction of speakerphone singing
View Patent ↗In a speakerphone device identification of signals (i.e., voice input or speaker output) in a process for reducing acoustic feedback, in a communication device, is accomplished by adding a signature noise (i.e., an identification mark) to output signals radiated by the speaker to enable these signals to be separated from speech input to the microphone. Having identified the signal (i.e., speech output) likely to cause a “singing” phenomenon, appropriate insertion loss to reduce the feedback may be added to the appropriate speech path within the communication device, to reduce a probability of singing.
1. A method of reducing feedback in a communication device having a microphone and a speaker, the communication device being connected to a network; the method comprising:
applying an identification mark to output from the speaker;
correlating the identification mark with a signal from the microphone to produce a correlation signal reflective of an amount of feedback;
introducing insertion loss in the microphone and speaker based on the correlation signal and relative signal levels of the microphone and speaker, in order to reduce the feedback.
2. The method of claim 1 wherein the relative signal levels of the microphone and speaker are determined by performing:
detecting an envelope of the signal from the microphone; and
detecting an envelope of a signal from the network.
3. The method of claim 2 wherein introducing insertion loss includes:
introducing the insertion loss based on a comparison of the signal envelopes.
4. The method of claim 1 wherein introducing insertion loss includes:
introducing the insertion loss in direct proportion to the correlation signal.
5. The method of claim 1 wherein applying an identification mark to output from the speaker includes:
combining the speaker output with a pseudo-noise (PN) signal waveform.
6. The method of claim 5 , wherein combining the speaker output includes modulating the PN signal with an envelope of the speech signal fed to the speaker.
7. The method of claim 1 , further comprising:
applying a second identification mark to output from the speaker; and
adjusting echo canceling circuitry using the second identification mark.
8. The method of claim 1 , further comprising:
filtering the correlation signal to band limit the signal.
9. The method of claim 1 , wherein introducing insertion loss comprises transmitting loss control signals to gain cells.
10. The method of claim 1 , wherein introducing insertion loss is effective to prevent buildup of feedback where neither incoming nor outgoing speech is present.
11. The method of claim 1 , wherein the identification mark is a PN sequence using several frequency sub-bands, each with a PN sequence to adjust switched loss in each of the bands.
12. The method of claim 11 , further comprising:
introducing a second PN sequence orthogonal to the PN sequence for adjusting an operation of an echo canceller.
13. A speakerphone connected to a communication network, comprising:
a loudspeaker for providing voice output from an output path;
a microphone for accepting voice input to an input path;
a signal identification mark generator connected to modulate a voice output signal with a signal identification mark;
a correlator connected to both the signal identification mark generator and the input path to produce a correlation signal reflective of an amount of feedback; and
a loss control processor to insert loss in the input and output paths responsive the correlation signal, to reduce signal feedback between the input and output path.
14. The speakerphone of claim 13 further comprising:
an output signal envelope detector connected to the output path; and
an input signal envelope detector connected to the input path;
wherein the loss control processor is further responsive to relative signal envelope sizes of the input and output paths.
15. The speakerphone of claim 14 , wherein the output signal envelope detector is further connected to a gain cell for receiving the signal identification mark and modulating the signal identification mark with an envelope of the output signal.
16. The speakerphone of claim 13 further comprising:
a filter coupling the correlator to the loss control processor.
17. The speakerphone of claim 13 , wherein the signal identification mark generator is a PN signal generator.