Granted Patent B2

US 12,482,467 · App. 18/598,339 · Granted Nov 25, 2025

Inventors: Anthony John Wood (San Jose, CA); David Stern (Los Gatos, CA); Gregory Mack Garner (Springdale, AZ)

Assignee: Roku, Inc.

G10L15/22G06F3/167H04L67/1014G10L2015/088G10L2015/223

Abstract

Claims (45)

1 . A computer-implemented method for providing audio control using multiple digital assistants, comprising:

selecting, by an audio platform, a first digital assistant from a plurality of digital assistants in the audio platform to process an audio input using a trigger word in the audio input, wherein the selected first digital assistant is mapped to the trigger word;

determining, by the audio platform, that a second digital assistant from the plurality of digital assistants in the audio platform to process an intent associated with the audio input more often than the selected first digital assistant based on tracking of the audio input, wherein the tracking comprises determining a time of day and location of the audio input; and

selecting, by the audio platform, the second digital assistant from the plurality of digital assistants in the audio platform to process the audio input based on the determining.

2 . The computer-implemented method of claim 1 , further comprising:

transmitting the intent to a voice adaptor at a media device, wherein the voice adaptor selects an application to process the intent based on a fixed rule, a default application setting, a search result, or metadata in the intent.

3 . The computer-implemented method of claim 1 , further comprising:

refining the intent based on information in a cloud computing platform.

4 . The computer-implemented method of claim 1 , further comprising:

generating the intent from the audio input using the selected first digital assistant.

5 . The computer-implemented method of claim 1 , further comprising:

converting the audio input into a text input using an automated speech recognizer associated with the selected first digital assistant; and

generating the intent from the text input using a natural language unit associated with the selected first digital assistant.

6 . The computer-implemented method of claim 1 , wherein the determining further comprises:

determining that the second digital assistant processes the intent associated with the audio input more often than the selected first digital assistant based on crowdsourced data, wherein the crowdsourced data indicates how often each digital assistant in the plurality of digital assistants is used to process the intent.

7 . The computer-implemented method of claim 6 , further comprising:

in response to selecting the second digital assistant, incrementing a count in the crowdsourced data that indicates a number of times the second digital assistant was selected.

8 . An audio platform, comprising:

a memory; and

at least one processor coupled to the memory and configured to:

select a first digital assistant from a plurality of digital assistants in the audio platform to process audio input using a trigger word in the audio input, wherein the selected first digital assistant is mapped to the trigger word;

determine that a second digital assistant from the plurality of digital assistants in the audio platform to process an intent associated with the audio input more often than the selected first digital assistant based on tracking of the audio input, wherein the tracking comprises determining a time of day and location of the audio input; and

select the second digital assistant from the plurality of digital assistants in the audio platform to process the audio input based on the determining.

9 . The audio platform of claim 8 , wherein the at least one processor is further configured to:

transmit the intent to an audio adaptor at a media device, wherein the audio adaptor selects an application to process the intent based on a fixed rule, a default application setting, a search result, or metadata in the intent.

10 . The audio platform of claim 8 , wherein the at least one processor is further configured to:

refine the intent based on information in a cloud computing platform.

11 . The audio platform of claim 8 , wherein the at least one processor is further configured to:

generate the intent from the audio input using the selected first digital assistant.

12 . The audio platform of claim 8 , wherein the at least one processor is further configured to:

convert the audio input into a text input using an automated speech recognizer associated with the selected first digital assistant; and

generate the intent from the text input using a natural language unit associated with the selected first digital assistant.

13 . The audio platform of claim 8 , wherein to determine that the second digital assistant processes the intent associated with the audio input more often than the selected first digital assistant, the at least one processor is further configured to:

determine that the second digital assistant processes the intent associated with the audio input more often than the selected first digital assistant based on crowdsourced data, wherein the crowdsourced data indicates how often each digital assistant in the plurality of digital assistants is used to process a type of the intent.

14 . The audio platform of claim 13 , wherein the at least one processor is further configured to:

in response to selecting the second digital assistant, increment a count in the crowdsourced data that indicates a number of times the second digital assistant was selected.

15 . A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device of a command module, cause the at least one computing device to perform operations comprising:

transmitting an audio input to an audio platform, wherein the audio platform selects a first digital assistant from a plurality of digital assistants in the audio platform to process the audio input using a trigger word in the audio input, determines that a second digital assistant from the plurality of digital assistants in the audio platform to process an intent associated with the audio input more often than the selected first digital assistant based on tracking of the audio input, wherein the tracking comprises determining a time of day and location of the audio input, and selects the second digital assistant from the plurality of digital assistants in the audio platform to process the audio input based on the determining; and

receiving the intent from the audio platform.

16 . The non-transitory computer-readable medium of claim 15 , wherein the receiving the intent from the audio platform further comprises:

receiving the intent at an audio adaptor, wherein the audio adaptor selects an application to process the intent based on a fixed rule, a default application setting, a search result, or metadata in the intent.

17 . The non-transitory computer-readable medium of claim 15 , wherein the audio platform refines the intent based on information in a cloud computing platform.

18 . The non-transitory computer-readable medium of claim 15 , wherein the audio platform converts the audio input into a text input using an automated speech recognizer associated with the selected first digital assistant, and generates the intent from the text input using a natural language unit associated with the selected first digital assistant.

19 . The non-transitory computer-readable medium of claim 15 , wherein the audio platform determines that the second digital assistant processes the intent associated with the audio input more often than the selected first digital assistant based on crowdsourced data, wherein the crowdsourced data indicates how often each digital assistant in the plurality of digital assistants is used to process the intent.

20 . The non-transitory computer-readable medium of claim 19 , wherein the audio platform, in response to selecting the second digital assistant, increments a count in the crowdsourced data that indicates a number of times the second digital assistant was selected.

Assignments (2)

Continuity (5)

References Cited (221)