Echo control in binaural adaptive noise cancellation systems in headsets
First Claim
1. A method of enhancing speech of a headphone user, the method comprising:
- receiving a primary signal derived from at least one microphone associated with the headphone, the primary signal configured to include a component of speech from the user;
generating an echo reference signal by pre-filtering at least one playback signal, the at least one playback signal provided by an audio source to be rendered into an acoustic signal by an acoustic driver associated with the headphone;
receiving a noise reference signal representative of acoustic noise in the environment of the headphone; and
filtering the primary signal to reduce components correlated to the noise reference signal and to reduce components correlated to the echo reference signal, to provide a voice estimate signal.
1 Assignment
0 Petitions
Accused Products
Abstract
A headphone, headphone system, and method is provided to enhance speech pick-up from a user of a headphone, by removing noise and echo components. A primary signal is derived from at least one microphone associated with the headphone. The primary signal is configured to include a component of speech from the user. A noise reference signal is representative of acoustic noise in the environment of the headphone. A playback signal is provided by an audio source to be rendered by an acoustic driver associated with the headphone, and an echo reference signal is representative of the playback signal. The systems and methods filter the primary signal to reduce noise and echo components, based on the noise reference signal and the echo reference signal, respectively, to provide a voice estimate signal.
68 Citations
18 Claims
-
1. A method of enhancing speech of a headphone user, the method comprising:
-
receiving a primary signal derived from at least one microphone associated with the headphone, the primary signal configured to include a component of speech from the user; generating an echo reference signal by pre-filtering at least one playback signal, the at least one playback signal provided by an audio source to be rendered into an acoustic signal by an acoustic driver associated with the headphone; receiving a noise reference signal representative of acoustic noise in the environment of the headphone; and filtering the primary signal to reduce components correlated to the noise reference signal and to reduce components correlated to the echo reference signal, to provide a voice estimate signal. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. An audio system, comprising:
-
at least one left microphone coupled to a left earpiece to provide a left primary signal; at least one right microphone coupled to a right earpiece to provide a right primary signal; and a signal processor configured to; receive at least one playback signal, derive an echo reference signal by pre-filtering the at least one playback signal, provide a combined signal from the left and right primary signals, provide a noise reference signal from the left and right primary signals, and filter the combined primary signal to reduce components correlated to the noise reference signal and to reduce components correlated to the echo reference signal, to provide a voice estimate signal. - View Dependent Claims (11, 12, 13, 14)
-
-
15. A headphone comprising:
-
at least one microphone coupled to an earpiece to provide a microphone signal; at least one input to receive a playback signal from an audio source; at least one acoustic driver coupled to the earpiece and configured to convert the playback signal into an acoustic signal; a noise reduction filter configured to receive a noise reference signal derived from the microphone signal and to provide an estimated noise signal; a pre-filter configured to pre-filter the playback signal to derive an echo reference signal; an echo cancellation filter configured to receive the echo reference signal derived from the playback signal and to provide an estimated echo signal; and a signal processor configured to provide an estimated voice signal derived at least in part from the microphone signal, including subtracting the estimated noise signal and the estimated echo signal. - View Dependent Claims (16, 17, 18)
-
Specification