# US 12,183,337 B2 · Voice command trigger words > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,183,337 B2 - Document ID: US12183337B2 - Status: Granted - Kind: B2 - Application number: 17/380,782 - Filed: 2021-07-20 - Published: 2022-03-17 - Granted: 2024-12-31 - Earliest claimed: 2018-12-06 - Assignees: Comcast Cable Communications, LLC - Applicants: Comcast Cable Communications, LLC - Inventors: Michael Chen - Web page: https://patentgrep.com/patent/US12183337B2 ## CPC - G10L 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue - G10L 15/08: Speech classification or search - G10L 15/30: Distributed recognition, e.g. in client-server systems, for mobile phones or network applications - G06F 1/325: {Power saving in peripheral device} - G10L 2015/088: {Word spotting} - G10L 2015/223: {Execution procedure of a spoken command} - G10L 2015/228: {of application context} - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract Methods, systems, and apparatuses are described for context-based speech processing. A current context may be determined. A subset of trigger words or phrases may be selected, from a plurality of trigger words, as trigger words that a computing device is configured to recognize in the determined current context. The computing device may be controlled to listen for only the subset of trigger words during speech recognition. ## Claims ### Claim 1 A method comprising: determining, by a computing device, information indicating content being output by another device in an environment associated with the computing device; performing speech recognition using a subset of trigger commands, of a plurality of trigger commands, that are based on the indicated content; and causing, based on a received audio input that comprises at least one trigger command of the subset of trigger commands, performance of an action. ### Claim 2 (depends on 1) The method of claim 1, wherein the another device is physically separate from the computing device. ### Claim 3 (depends on 1) The method of claim 1, wherein the subset of trigger commands is further based on a distance of the computing device from a controllable device. ### Claim 4 (depends on 1) The method of claim 1, wherein the subset of trigger commands is further based on detection of an individual. ### Claim 5 (depends on 1) The method of claim 1, further comprising: determining, based on the indicated content, one or more devices, wherein the causing the performance of the action comprises causing at least one of the one or more devices to perform the action. ### Claim 6 (depends on 1) The method of claim 1, further comprising: determining a difference between energy usage associated with listening for the subset of trigger commands and energy usage associated with listening for the plurality of trigger commands; and based on a determination that the difference satisfies a threshold, listening for the subset of trigger commands. ### Claim 7 (depends on 1) The method of claim 1, wherein the information indicating the content comprises information indicating a title of the content. ### Claim 8 (depends on 1) The method of claim 1, further comprising receiving, from the another device, the information indicating the content. ### Claim 9 (depends on 1) The method of claim 1, wherein the performing speech recognition using the subset of trigger commands is further based on an energy usage associated with listening for the subset of trigger commands. ### Claim 10 (depends on 1) The method of claim 1, further comprising: determining data identifying a plurality of combinations of states of a plurality of devices, wherein each combination is associated with a corresponding subset of trigger commands; determining a combination of states of the plurality of devices; and selecting, based on the data and the combination of states of the plurality of devices, the subset of trigger commands. ### Claim 11 A method comprising: determining, by a computing device in an environment comprising one or more devices, based on a time of day and audio output by another device in the environment, a set of trigger commands, wherein the set of trigger commands comprises fewer than a quantity of trigger commands in a plurality of trigger commands, and wherein at least one trigger command, of the set of trigger commands, is associated with the another device; and causing performance of speech recognition based on the set of trigger commands. ### Claim 12 (depends on 11) The method of claim 11, wherein the audio output in the environment comprises audio associated with one or more of: a doorbell; or an alarm. ### Claim 13 (depends on 11) The method of claim 11, wherein the determining the set of trigger commands is further based on data received from one or more of: a window or door sensor; a proximity sensor; a grip sensor; or a vehicle seat sensor. ### Claim 14 (depends on 11) The method of claim 11, wherein the determining the set of trigger commands is further based on data indicating one or more of: a vehicle seat being occupied; a movement of a device receiving a speech input; a cellular signal strength of one device of the one or more devices; or a power level of at least one device of the one or more devices. ### Claim 15 (depends on 11) The method of claim 11, further comprising: receiving an indication that a result of the speech recognition on at least a portion of a speech input does not meet a threshold; performing additional speech processing on the at least the portion of the speech input; and causing, based on a result of the additional speech processing, performance of an action. ### Claim 16 (depends on 11) The method of claim 11, further comprising: identifying, based on information received from at least one device, an identity of a person providing a speech input, wherein causing performance of the speech recognition comprises causing performance of the speech recognition on the speech input and further based on a user profile associated with the identity of the person. ### Claim 17 An apparatus comprising: one or more processors; and memory storing instructions that, when executed by the one or more processors, cause the apparatus to: determine information indicating content being output by another device in an environment associated with the apparatus; perform speech recognition using a subset of trigger commands, of a plurality of trigger commands, that are based on the indicated content; and cause, based on a received audio input that comprises at least one trigger command of the subset of trigger commands, performance of an action. ### Claim 18 (depends on 17) The apparatus of claim 17, wherein the another device is physically separate from the apparatus. ### Claim 19 (depends on 17) The apparatus of claim 17, wherein the subset of trigger commands are further based on a distance of the apparatus from a controllable device. ### Claim 20 (depends on 17) The apparatus of claim 17, wherein the subset of trigger commands are further based on detection of an individual. ### Claim 21 (depends on 17) The apparatus of claim 17, wherein the instructions, when executed by the one or more processors, further cause the apparatus to: determine, based on the indicated content, one or more devices; and cause the performance of the action by causing at least one of the one or more devices to perform the action. ### Claim 22 (depends on 17) The apparatus of claim 17, wherein the instructions, when executed by the one or more processors, cause the apparatus to: determine a difference between energy usage associated with listening for the subset of trigger commands and energy usage associated with listening for the plurality of trigger commands; and based on a determination that the difference satisfies a threshold, listen for the subset of trigger commands. ### Claim 23 An apparatus comprising: one or more processors; and a memory storing computer-readable instructions that, when executed by the one or more processors, cause the apparatus to: determine, in an environment comprising one or more devices, based on time of day and audio output by another device, a set of trigger commands, wherein the set of trigger commands comprises fewer than a quantity of trigger commands in a plurality of trigger commands, and wherein at least one trigger command, of the set of trigger commands, is associated with the another device; and cause performance of speech recognition based on the set of trigger commands. ### Claim 24 (depends on 23) The apparatus of claim 23, wherein the audio output in the environment comprises audio associated with one or more of: a doorbell; or an alarm. ### Claim 25 (depends on 23) The apparatus of claim 23, wherein the instructions, when executed by the one or more processors, cause the apparatus to determine the set of trigger commands further based on data received from one or more of: a window or door sensor; a proximity sensor; a grip sensor; or a vehicle seat sensor. ### Claim 26 (depends on 23) The apparatus of claim 23, wherein the instructions, when executed by the one or more processors, cause the apparatus to determine the set of trigger commands further based on data indicating one or more of: a vehicle seat being occupied; a movement of a device receiving a speech input; a cellular signal strength of one device of the one or more devices; or a power level of at least one device of the one or more devices. ### Claim 27 (depends on 23) The apparatus of claim 23, wherein the instructions, when executed by the one or more processors, cause the apparatus to: receive an indication that a result of the speech recognition on at least a portion of a speech input does not meet a threshold; perform additional speech processing on the at least the portion of the speech input; and cause, based on a result of the additional speech processing, performance of an action. ### Claim 28 (depends on 23) The apparatus of claim 23, wherein the instructions, when executed by the one or more processors, cause the apparatus to: identify, based on information received from at least one device of the one or more devices, an identity of a person providing a speech input, and cause the performance of the speech recognition on the speech input and further based on a user profile associated with the identity of the person. ### Claim 29 (depends on 1) The method of claim 1, further comprising detecting, using one or more sensors, the content being output by the another device. ### Claim 30 A non-transitory computer-readable medium storing instructions that, when executed, configure a computing device to: determine information indicating content being output by another device in an environment associated with the computing device; perform speech recognition using a subset of trigger commands, of a plurality of trigger commands, that are based on the indicated content; and cause, based on a received audio input that comprises at least one trigger command of the subset of trigger commands, performance of an action. ### Claim 31 (depends on 30) The non-transitory computer-readable medium of claim 30, wherein the another device is physically separate from the computing device. ### Claim 32 (depends on 30) The non-transitory computer-readable medium of claim 30, wherein the subset of trigger commands are further based on a distance of the computing device from a controllable device. ### Claim 33 (depends on 30) The non-transitory computer-readable medium of claim 30, wherein the subset of trigger commands are further based on detection of an individual. ### Claim 34 (depends on 30) The non-transitory computer-readable medium of claim 30, wherein the instructions, when executed, further configure the computing device to: determine, based on the indicated content, one or more devices; and cause the performance of the action by causing at least one of the one or more devices to perform the action. ### Claim 35 (depends on 30) The non-transitory computer-readable medium of claim 30, wherein the instructions, when executed, configure the computing device to: determine a difference between energy usage associated with listening for the subset of trigger commands and energy usage associated with listening for the plurality of trigger commands; and based on a determination that the difference satisfies a threshold, listen for the subset of trigger commands. ### Claim 36 A non-transitory computer-readable medium storing instructions that, when executed, configure a computing device to: determine, in an environment comprising one or more devices, based on time of day and audio output by another device, a set of trigger commands, wherein the set of trigger commands comprises fewer than a quantity of trigger commands in a plurality of trigger commands, and wherein at least one trigger command, of the set of trigger commands, is associated with the another device; and cause performance of speech recognition based on the set of trigger commands. ### Claim 37 (depends on 36) The non-transitory computer-readable medium of claim 36, wherein the audio output in the environment comprises audio associated with one or more of: a doorbell; or an alarm. ### Claim 38 (depends on 36) The non-transitory computer-readable medium of claim 36, wherein the instructions, when executed, further configure the computing device to: receive an indication that a result of the speech recognition on at least a portion of a speech input does not meet a threshold; perform additional speech processing on the at least the portion of the speech input; and cause, based on a result of the additional speech processing, performance of an action. ### Claim 39 (depends on 36) The non-transitory computer-readable medium of claim 36, wherein the instructions, when executed, further configure the computing device to: identify, based on information received from at least one device of the one or more devices, an identity of a person providing a speech input, and cause the performance of the speech recognition on the speech input and further based on a user profile associated with the identity of the person.