US 10,219,099 B1Granted

System for compensation of presentation by wireless devices

AssigneeAmazon Technologies, Inc.

InventorMichael Alan Pogue

Life of this patent

Estimating the term…

Loading patent…

Abstract

A group of devices connected on a network operate in conjunction with one another to present content. Individual devices may be wirelessly connected to presentation devices such as speakers. Skew is a delay between when content is sent to the presentation device and then output. Drift is a difference in timing of output between two or more presentation devices not synchronized to a common clock. Non-zero values of skew and drift may result in poor content presentation by the group. An actual rate at which data is transferred to a presentation device is measured and compared to an expected rate. This difference is analyzed to determine a drift value. The content to be delivered to the presentation device is resampled. During presentation the combination of the drift and the resampled content reduces skew. The drift value may also be used to lookup a skew value used to delay initial presentation.

Claims20

Dependent claims

A first audio device comprising:

a first wireless local area network (WLAN) interface;
a first Bluetooth interface coupled to a first wireless speaker;
a first memory storing first computer-executable instructions;
a first hardware processor to execute the first computer-executable instructions to:
determine an expected number of samples of audio data per unit time;
send, at a first time, first audio data to the first wireless speaker, wherein the first audio data is sampled at a first sample rate;
determine, for a plurality of time intervals between the first time and a second time, an actual number of samples of the first audio data sent per unit time to the first wireless speaker;
determine, for a portion of the plurality of time intervals, a first set of residuals by calculating a difference between the expected number of samples per unit time and the actual number of samples sent per unit time for a particular one of the portion of the plurality of time intervals;
determine a first regression line using a linear regression of the first set of residuals;
determine, based on a slope of the first regression line, a drift value;
retrieve, based on the drift value, a skew value from previously stored data, wherein the skew value is indicative of an initial delay between sending audio data to the first wireless speaker and output of corresponding audio by the first wireless speaker;
determine an adjusted number of samples per unit time by summing an additive inverse of the drift value to the expected number of samples per unit time;
and
send, at a third time to the first wireless speaker, second audio data that includes a presentation delay with a length of the skew value and wherein the second audio data is sent at the adjusted number of samples per unit time.

11 more elements

Dependencies are inferred from the claim text.