Back to results
US 12,615,472 B2Granted
Activity detection using a hearing instrument
AssigneeStarkey Laboratories, Inc.
InventorsVishal Vijayakumar, Matthew D. Kleffner, Tao Zhang
Life of this patent
Estimating the term…
Abstract
A computing system includes a memory and at least one processor. The memory is configured to store motion data indicative of motion of a hearing instrument. The at least one processor is configured to determine a type of activity performed by a user of the hearing instrument and output data indicating the type of activity performed by the user.
Claims20
A hearing instrument comprising:
- a memory configured to store a plurality of activity models, wherein each of the activity models is machine trained to determine a respective type of activity that is performed by a user of the hearing instrument;
- and
- at least one processor configured to:
- apply one or more activity models in a hierarchy of the activity models to motion data at least until a particular activity model of the plurality of activity models detects that the user is performing a particular type of activity that the particular activity model is trained to detect or until all of the activity models have been applied to the motion data and have generated a negative output, wherein the motion data is indicative of motion of the hearing instrument, and
- responsive to the particular activity model determining that the user is performing the particular type of activity, output data indicating the particular type of activity is a type of activity being performed by the user.
2 more elements
A method comprising:
- receiving, by at least one processor, motion data indicative of motion of a hearing instrument;
- applying, by the at least one processor, one or more activity models in a hierarchy of a plurality of activity models to the motion data at least until a particular activity model of the plurality of activity models detects that a user of the hearing instrument is performing a particular type of activity that the particular activity model is trained to detect or until all of the activity models have been applied to the motion data and have generated a negative output, wherein each of the activity models is machine trained to determine a respective type of activity that is performed by the user;
- and
- responsive to the particular activity model determining that the user is performing the particular type of activity, outputting data indicating the particular type of activity is a type of activity being performed by the user.
1 more element
A non-transitory computer-readable storage medium comprising instructions that, when executed by at least one processor, cause the at least one processor to:
- receive motion data indicative of motion of a hearing instrument;
- apply one or more activity models in a hierarchy of a plurality of activity models to the motion data at least until a particular activity model of the plurality of activity models detects that a user of the hearing instrument is performing a particular type of activity that the particular activity model is trained to detect or until all of the activity models have been applied to the motion data and have generated a negative output, wherein each of the activity models is machine trained to determine a respective type of activity that is performed by the user;
- and
- responsive to the particular activity model determining that the user is performing the particular type of activity, output data indicating the particular type of activity is a type of activity being performed by the user.
1 more element
Dependencies are inferred from the claim text.