Back to results
US 11,893,324 B2Granted
Ear-wearable device shell modeling
AssigneeStarkey Laboratories, Inc.
InventorTao Zhang
Life of this patent
Estimating the term…
Abstract
A computing device is described that obtains a representation of a target ear canal of a user. Using a machine-learned model that has been trained based at least in part on representations of previously fabricated ear-wearable devices, the computing device generates a representation of an ear-wearable device for the target ear canal.
Claims23
A method comprising:
- training a machine learning (ML) model to generate a representation of an ear-wearable device for a target ear canal, wherein training the ML model comprises:
- receiving a data set that includes a representation of a historic ear impression and a corresponding target representation of an ear-wearable device;
- providing the historic ear impression to the ML model, wherein the ML model is configured to output a generated representation of the ear-wearable device;
- applying a cost function to the output of the ML model to determine a cost, wherein the cost function includes a weighted mean square error of deviations of the generated representation of the ear-wearable device and the target representation of the ear-wearable device;
- and
- updating model parameters of the ML model based on the cost;
- receiving, by at least one processor, a representation of a target ear canal of a patient;
- and
- applying, by the at least one processor, the ML model to generate a representation of an ear-wearable device for the target ear canal, wherein input to the ML model includes the representation of the target ear canal of the patient.
6 more elements
A non-transitory computer-readable storage medium comprising instructions that when executed by at least one processor, cause the at least one processor to:
- train a machine learning (ML) model to generate a representation of an ear-wearable device for a target ear canal, wherein the at least one processor is configured to, as part of training the ML model:
- receiving a data set that includes a representation of a historic ear impression and a corresponding target representation of an ear-wearable device;
- provide the historic ear impression to the ML model, wherein the ML model is configured to output a generated representation of the ear-wearable device;
- apply a cost function to the output of the ML model to determine a cost, wherein the cost function includes a weighted mean square error of deviations of the generated representation of the ear-wearable device and the target representation of the ear-wearable device;
- and
- update model parameters of the ML model based on the cost;
- receive a representation of a target ear canal of a patient;
- and
- apply the ML model to generate a representation of an ear-wearable device for the target ear canal, wherein input to the ML model includes the representation of the target ear canal of the patient.
6 more elements
A computing device comprising:
- one or more data storage devices configured to store a machine-learned (ML) model;
- and
- one or more processors configured to:
- train the ML model to generate a representation of an ear-wearable device for a target ear canal, wherein train the ML model comprises:
- receive a data set that includes a representation of a historic ear impression and a corresponding target representation of an ear-wearable device;
- provide the historic ear impression to the ML model, wherein the ML model is configured to output a generated representation of the ear-wearable device:
- apply a cost function to the output of the ML model to determine a cost, wherein the cost function includes a weighted mean square error of deviations of the generated representation of the ear-wearable device and the target representation of the ear-wearable device;
- and
- update model parameters of the ML model based on the cost;
- receive a representation of a target ear canal of a user;
- and
- apply a machine-learned model to generate a representation of an ear-wearable device for the target ear canal, wherein input to the machine-learned model includes the representation of the target ear canal of the user and representations of one or more internal components of the ear-wearable device.
9 more elements
Dependencies are inferred from the claim text.