# US 10,313,845 B2 · Proactive speech detection and alerting > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,313,845 B2 - Document ID: US10313845B2 - Status: Granted - Kind: B2 - Application number: 15/615,607 - Filed: 2017-06-06 - Published: 2018-12-06 - Granted: 2019-06-04 - Earliest claimed: 2017-06-06 - Assignees: Microsoft Technology Licensing, LLC - Applicants: Microsoft Technology Licensing, LLC - Inventors: Anirudh Koul; Venkata Naga Vijaya Swetha Machanavajhala; Stéphane Morichère-Matte; David Dai Wei Zhang; Anass Al-Wohoush; Jan Ervin Mickel Takata Clarin; Sheng-Ting Lin; Sitthinon Jinaphant; Shery Sharonjit Sumal - Web page: https://patentgrep.com/patent/US10313845B2 ## CPC - H04W 4/12: Messaging; Mailboxes; Announcements - G10L 15/08: Speech classification or search - G10L 15/26: Speech to text systems (G10L15/08 takes precedence) - G10L 15/30: Distributed recognition, e.g. in client-server systems, for mobile phones or network applications - G10L 25/84: for discriminating voice from noise - H04M 1/72421: with automatic activation of emergency service functions, e.g. upon sensing an alarm - H04M 2250/74: with voice recognition means - H04W: WIRELESS COMMUNICATION NETWORKS (broadcast communication H04H; communication systems using wireless links for non-selective communication, e.g. wireless extensions H04M1/72) - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING - H04M: TELEPHONIC COMMUNICATION (circuits for controlling other apparatus via a telephone cable and not involving telephone switching apparatus G08) ## Abstract Non-limiting examples of the present disclosure describe proactive speech detection on behalf of a user and alerting the user when a specific word, name, etc. is detected. Speech detection is actively executed through a computing device, where the speech detection analyzes spoken utterances in association with a dynamic grammar file stored locally on the computing device. An alert is generated that indicates when a spoken word of the utterances matches a word stored in the dynamic grammar file. The alert may be displayed, for example, through the computing device. The alert provides indication that the spoken word is identified in the spoken utterances. In further examples, a buffered window of the spoken utterances is captured which is associated with a detection of the specific word. A live transcription of the content in the buffered window is generated and provided to a computing device of the user. ## Claims ### Claim 1 A method comprising: receiving, through a user computing device, a user input identifying a word for a user-defined grammar file; dynamically updating the user-defined grammar file to comprise the word; actively executing, through the computing device, speech detection that analyzes a spoken utterance in association with the user-defined grammar file stored locally on the computing device; detecting that a spoken word of the spoken utterance matches the word stored in the user-defined grammar file; generating, based on the detecting, an alert that indicates that the word is identified in the spoken utterance; and displaying, through the computing device, the alert. ### Claim 2 (depends on 1) The method of claim 1, wherein the speech detection further comprises filtering out non-speech noise and silence from the spoken utterance and buffering the spoken utterance through a moving window buffer that continuously tracks intervals of sound utterances. ### Claim 3 (depends on 2) The method of claim 2, further comprising: providing, to a distributed speech recognition service, a buffered window corresponding with an interval of the sounds utterances that is associated with a detection of the word, receiving, from the distributed speech recognition service, a live transcription of content in the buffered window and displaying, through the computing device, a message that comprises the live transcription. ### Claim 4 (depends on 2) The method of claim 2, further comprising: analyzing, through an application executing on the computing device, a buffered window corresponding with an interval of the sounds utterances that is associated with a detection of the word, generating a live transcription for content of the buffered window, and displaying, through the computing device, a message that comprises the live transcription. ### Claim 5 (depends on 1) The method of claim 1, further comprising: providing a live stream of the spoken utterance in association with the displayed alert. ### Claim 6 (depends on 1) The method of claim 1, further comprising: presenting, in association with the alert, one or more selected from a group consisting of: a vibration of the computing device and an audio alert output through the computing device. ### Claim 7 (depends on 1) The method of claim 1, wherein the receiving of the user input comprises receiving, through a user interface, one or more spoken utterances for the word, and wherein the detecting of a matching between the spoken word and the word stored in the user-defined grammar file comprises matching audio data for the spoken word with audio data of the one or more spoken utterances for the word. ### Claim 8 (depends on 7) The method of claim 7, wherein the receiving of the user input comprises receiving, through a user interface, one or more spoken utterances for the word, and wherein the detecting of a matching between the spoken word and the word stored in the user-defined grammar file comprises matching audio data for the spoken word with audio data of the one or more spoken utterances for the word. ### Claim 9 A system comprising: at least one processor; and a memory, operatively connected with the at least one processor, storing computer-executable instructions that, when executed by the at least one processor, causes the at least one processor to execute a method that comprises: receiving, through a user computing device, a user input identifying a word for a user-defined grammar file; dynamically updating the user-defined grammar file to comprise the word; actively executing, through the computing device, speech detection that analyzes a spoken utterance in association with the user-defined grammar file stored locally on the computing device; detecting that a spoken word of the spoken utterance matches the word stored in the user-defined grammar file; generating, based on the detecting, an alert that indicates that the word is identified in the spoken utterance; and displaying, through the computing device, the alert. ### Claim 10 (depends on 9) The system of claim 9, wherein the speech detection further comprises filtering out non-speech noise and silence from the spoken utterance and buffering the spoken utterance through a moving window buffer that continuously tracks intervals of sound utterances. ### Claim 11 (depends on 10) The system of claim 10, wherein the method, executed by the at least one processor, further comprises: providing, to a distributed speech recognition service, a buffered window corresponding with an interval of the sounds utterances that is associated with a detection of the word, receiving, from the distributed speech recognition service, a live transcription of content in the buffered window and displaying, through the system, a message that comprises the live transcription. ### Claim 12 (depends on 10) The system of claim 10, wherein the method, executed by the at least one processor, further comprises: analyzing a buffered window corresponding with an interval of the sounds utterances that is associated with a detection of the word, generating a live transcription for content of the buffered window, and displaying a message that comprises the live transcription. ### Claim 13 (depends on 9) The system of claim 9, wherein the method, executed by the at least one processor, further comprises: providing a live stream of the spoken utterance in association with the displayed alert. ### Claim 14 (depends on 9) The system of claim 9, wherein the method, executed by the at least one processor, further comprises: presenting, in association with the alert, one or more selected from a group consisting of: a vibration of a component associated with the system and an audio alert output through the system. ### Claim 15 A computer storage media computer-executable instructions that, when executed by at least one processor, causes the at least one processor to execute a method comprising: receiving, through a user computing device, a user input identifying a word for a user-defined grammar file; dynamically updating the user-defined grammar file to comprise the word; actively executing, through the computing device, speech detection that analyzes a spoken utterance in association with the user-defined grammar file stored locally on the computing device; detecting that a spoken word of the spoken utterance matches the word stored in the user-defined grammar file; generating, based on the detecting, an alert that indicates that the word is identified in the spoken utterance; and displaying, through the computing device, the alert. ### Claim 16 (depends on 15) The computer storage media of claim 15, wherein the speech detection further comprises filtering out non-speech noise and silence from the spoken utterance and buffering the spoken utterance through a moving window buffer that continuously tracks intervals of sound utterances. ### Claim 17 (depends on 16) The computer storage media of claim 16, wherein the executed method further comprising: providing, to a distributed speech recognition service, a buffered window corresponding with an interval of the sounds utterance that is associated with a detection of the word, receiving, from the distributed speech recognition service, a live transcription of content in the buffered window and displaying, through the computing device, a message that comprises the live transcription. ### Claim 18 (depends on 16) The computer storage media of claim 16, wherein the executed method further comprising: analyzing, through an application executing on the computing device, a buffered window corresponding with an interval of the sounds utterances that is associated with a detection of the word, generating a live transcription for content of the buffered window, and displaying a message that comprises the live transcription. ### Claim 19 (depends on 15) The computer storage media of claim 15, wherein the executed method further comprising: providing a live stream of the spoken utterance in association with the displayed alert. ### Claim 20 (depends on 15) The computer storage media of claim 15, wherein the executed method further comprising: presenting, in association with the alert, one or more selected from a group consisting of: a vibration of the computing device and an audio alert output through the computing device.