Curating audio and IR commands through machine learning
AssigneeAmazon Technologies, Inc.
InventorsMichael L S Dodge, Charles Shearer Dorner
Life of this patent
Estimating the term…
Abstract
Techniques for curating audio and IR commands using machine learning may be provided. For example, the system can receive an audio stream that includes a plurality of audio segments provided by a particular user (e.g., as determined through acoustic fingerprinting) and the system can store the audio stream and/or segments in the user's profile. The system can also store a command associated with the audio segment(s). When a portion of the audio is provided in conjunction with a same command over a threshold number of repetitions, the next time that that audio segment is received, the system may provide the command that corresponds with that audio segment to an output device to cause an operation of the output device.
Claims20
A computer-implemented method, comprising:
- receiving an audio stream that includes a plurality of audio segments, the audio stream corresponding to an acoustic fingerprint associated with a user profile;
- obtaining, from the user profile, data indicative of a command received by an output device associated with the user profile;
- determining, from the user profile associated with the acoustic fingerprint, that an audio segment in the audio stream is associated in time with the command;
- when a number of repetitions of an association in time of the audio segment and command exceeds a threshold, updating the user profile with the association between the audio segment and the command;
- and
- in response to a subsequent receipt of the audio segment, providing the command that corresponds with the audio segment to the output device to cause an operation of the output device, the command provided based at least in part on the association between the audio segment and the command.
A computer-implemented method, comprising:
- receiving, by a computer system, an audio stream that includes a plurality of audio segments, the audio stream corresponding to an acoustic fingerprint associated with a user profile;
- obtaining, from the user profile, data indicative of a command received by an output device associated with the user profile;
- determining, from the user profile associated with the acoustic fingerprint, that an audio segment in the audio stream is associated in time with the command;
- when a number of repetitions of an association in time of the audio segment and command exceeds a threshold, updating the user profile with the association between the audio segment and the command;
- and
- in response to a subsequent receipt of the audio segment, providing the command that corresponds with the audio segment to the output device to cause an operation of the output device, the command provided based at least in part on the association between the audio segment and the command.
A computer system, comprising:
- a memory configured to store computer-executable instructions;
- and
- at least one processor in communication with the memory configured to execute the computer-executable instructions to at least:
- receive an audio stream that includes a plurality of audio segments, the audio stream corresponding to an acoustic fingerprint associated with a user profile;
- obtain, from the user profile, data indicative of a command received by an output device associated with the user profile;
- determine, from the user profile associated with the acoustic fingerprint, that an audio segment in the audio stream is associated in time with the command;
- when a number of repetitions of an association in time of the audio segment and command exceeds a threshold, update the user profile with the association between the audio segment and the command;
- and
- in response to a subsequent receipt of the audio segment, provide the command that corresponds with the audio segment to the output device to cause an operation of the output device, the command provided based at least in part on the association between the audio segment and the command.
Dependencies are inferred from the claim text.