Granted Patent B2

US 12,527,495 · App. 17/792,886 · Granted Jan 20, 2026

Inventors: Andrea Fanelli (Oakland, CA); Evan David Gitterman (Palo Alto, CA); Nathan Carl Swedlow (Oakland, CA); Alex Brandmeyer (Berkeley, CA); McGregor Steele Joyner (Alameda, CA); Poppy Anne Carrie Crum (Oakland, CA)

Assignee: Dolby Laboratories Licensing Corporation

A61B5/1114A61B5/256G06F3/013G06F3/015G06V40/18G06V40/70

Abstract

Claims (68)

1 . A system for determining a gaze direction of a user, comprising:

an audio end-point including a pair of earpieces, each earpiece intended to be worn adjacent to a respective one of the user's ears;

a set of earpiece electrodes arranged on each earpiece, such that the earpiece electrodes, in use, are placed in contact with the user's skin;

each earpiece electrode comprising a patch of compressible and electrically conductive foam material, configured to provide an electrical connection with the skin; and

circuitry connected to the earpiece electrodes and configured to:

receive a set of voltage signals from the set of earpiece electrodes;

multiplex the voltage signals into an input signal;

remove a predicted central voltage, representing a forward gaze direction, from the input signal, to provide a detrended signal;

determine the gaze direction based on the detrended signal; and

determine a subsequent predicted central voltage by:

calculating a current central voltage as a difference between the input signal and a voltage representing a current gaze direction;

providing a first estimate as a linear weighting of the current central voltage and a current estimated central voltage;

providing a second estimate as a mean of the current central voltage and a prediction based on a set of preceding predicted central voltages; and

determining the subsequent predicted central voltage as an average of the first and second estimates.

2 . The system in claim 1 , wherein the audio end-point is a pair of on-ear headphones, each on-ear headphone including a cuff configured to rest against the skin of a user's head, and wherein the earpiece electrodes are arranged on a rim of the cuff intended to rest against the skin.

3 . The system of claim 1 , wherein the conductive foam material has a thickness of less than one mm, and a vertical contact resistance of less than 0.005 Ω.

4 . The system of claim 1 , wherein the conductive foam material has a volume resistivity is in a range 1.3 to 3 Ω·cm.

5 . The system of claim 1 , wherein the conductive foam material has a surface resistivity of 0.1 to 0.12 Ω/sq.

6 . The system of claim 1 , wherein the prediction is made using an AR model and the set of predicted central voltages.

7 . The system of claim 1 , further comprising identifying a shift in the detrended signal representing a transition of the user's gaze from one saccade to another saccade.

8 . The system of claim 1 , further comprising determining a state of the gaze direction, the state being selected from a group including a left state and a right state, wherein the left and right states are defined by the detrended signal exceeding or falling below a defined threshold value.

9 . A method for determining a gaze direction of a user, comprising:

at an audio end-point including a pair of earpieces, each earpiece supporting a set of earpiece electrodes comprising a patch of compressible and electrically conducting material configured to provide an electrical connection with the user's skin, the audio end-point arranged on the user such that the earpiece electrodes are in contact with the user's skin in a vicinity of each respective ear, acquiring a set of voltage signals from the set of earpiece electrodes;

multiplexing the voltage signals into an input signal;

removing a predicted central voltage, representing a forward gaze direction, from the input signal, to provide a detrended signal; and

determining the gaze direction based on the detrended signal;

wherein the predicted central voltage is determined by:

calculating a current central voltage as a difference between the input signal and a voltage representing a current gaze direction;

providing a first estimate as a linear weighting of the current central voltage and a current estimated central voltage;

providing a second estimate as a mean of the current central voltage and a prediction based on a set of preceding predicted central voltages; and

determining the predicted central voltage as an average of the first and second estimates.

10 . The method of claim 9 , wherein the prediction is made using an AR model and the set of predicted central voltages.

11 . The method of claim 9 , further comprising identifying a step change in the detrended signal, the step change representing a transition of the user's gaze from one saccade to another saccade.

12 . The method of claim 9 , further comprising determining a state of the gaze direction, the state being selected from a group including a left state and a right state, wherein the left and right states are defined by the detrended signal exceeding or falling below a defined threshold value.

13 . The method of claim 9 , wherein the audio end-point is a pair of on-ear headphones, each on-ear headphone including a cuff configured to rest against the skin of a user's head, and wherein the earpiece electrodes are arranged on a rim of the cuff intended to rest against the skin.

14 . The system of claim 1 , further comprising:

at least one additional sensor; and

a neural network for estimating a direction of attention given input from the system and from the at least one additional sensor.

15 . The system according to claim 14 , wherein the at least one additional sensor includes at least one sensor for measuring a physiological response.

16 . The system according to claim 14 , wherein the additional sensor includes at least one of: an inertial measuring unit (IMU), a sensor for measuring heart rate, a sensor for measuring blood pressure, a temperature sensor, electrodes for electroencephalogram, EEG, and electromyogram, EMG.

17 . The system according to claim 14 , further comprising a system for improving auditory attention, comprising:

a blind source separation processor connected to receive the direction of attention and an audio mix including dialogue and noise, the blind source separation processor configured to extract dialogue received from the direction of attention; and

an amplifier for amplifying the extracted dialogue.

18 . The system according to claim 14 , further comprising a system for controlling a pass-through feature of noise-cancelling headphones, comprising:

a blind source separation processor connected to receive the direction of attention and an audio mix including dialogue and noise, the blind source separation processor configured to extract dialogue received from the direction of attention;

an amplifier for amplifying the extracted dialogue; and

a mixer for mixing the extracted dialogue into content intended for playback on the noise-cancelling headphones.

19 . A method for determining a gaze direction of a user, comprising:

receiving a set of voltage signals from a set of electrodes arranged on an audio endpoint worn by a user, the set of electrodes in contact with the user's skin, each electrode comprising a patch of compressible and electrically conducting material configured to provide an electrical connection with the user's skin;

multiplexing the voltage signals into an input signal;

calculating a current central voltage as a difference between the input signal and a voltage representing a current gaze direction;

providing a first estimate as a linear weighting of a current central voltage and a current estimated baseline voltage;

providing a second estimate as a mean of the current central voltage and a prediction based on a set of preceding predicted central voltages;

determining a predicted central voltage as an average of the first and second estimates;

removing the predicted central voltage from the input signal, to provide a detrended signal; and

determining the gaze direction based on the detrended signal.

20 . A system for determining a gaze direction of a user, comprising:

a set of electrodes arranged to be in contact with the user's skin, each electrode comprising a patch of compressible and electrically conducting material configured to provide an electrical connection with the skin; and

circuitry connected to the electrodes and configured to:

receive a set of voltage signals from the set of electrodes;

multiplex the voltage signals into an input signal;

calculating a current central voltage as a difference between the input signal and a voltage representing a current gaze direction;

provide a first estimate as a linear weighting of a current central voltage and a current estimated baseline voltage;

provide a second estimate as a mean of the current central voltage and a prediction based on a set of preceding predicted central voltages;

determine a predicted central voltage as an average of the first and second estimates;

remove the predicted central voltage from the input signal, to provide a detrended signal; and

determine the gaze direction based on the detrended signal.

21 . A non-transitory computer readable medium storing instructions which when executed by one or more processors of an electronic device, cause the electronic device to perform the method of claim 10 or claim 19 .

Assignments (2)

Priority Claims (1)

Continuity (3)

References Cited (44)