US 12,075,119 B2
Speaker-identification model for controlling operation of a media player
Frank Maker, Livermore, CA (US)
Assigned to Roku, Inc., San Jose, CA (US)
Filed by Roku, Inc., San Jose, CA (US)
Filed on Mar. 24, 2023, as Appl. No. 18/189,701.
Application 18/189,701 is a continuation of application No. 17/838,847, filed on Jun. 13, 2022, granted, now 11,641,505.
Prior Publication US 2023/0403430 A1, Dec. 14, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. H04N 21/45 (2011.01); H04N 21/422 (2011.01); H04N 21/4415 (2011.01); H04N 21/472 (2011.01)
CPC H04N 21/4532 (2013.01) [H04N 21/42203 (2013.01); H04N 21/4415 (2013.01); H04N 21/47202 (2013.01)] 17 Claims
OG exemplary drawing
 
9. A method performed by a media presentation system, the method comprising:
presenting an advertisement;
obtaining an audio signal, wherein the audio signal comprises a voice command and is obtained using a microphone of the media presentation system;
after presenting the advertisement, identifying, using the audio signal and a speaker-identification model, which of multiple speakers of a household uttered the voice command;
obtaining metadata for the identified speaker;
based on the identifying of the speaker using the audio signal and the speaker-identification model, generating an advertisement impression record that associates the metadata for the identified speaker with the advertisement;
obtaining another audio signal, wherein the other audio signal comprises another voice command;
determining, using the other audio signal and the speaker-identification model, that another speaker of the multiple speakers of the household uttered the voice command;
after obtaining the other audio signal, presenting the advertisement again; and
obtaining metadata for the other identified speaker; and generating another advertisement impression record that associates the metadata for the other identified speaker with the additional presentation of the advertisement.