# US 2019/0318010 A1 · Automated presentation control > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2019/0318010 A1 - Document ID: US20190318010A1 - Status: Published application - Kind: A1 - Application number: 15/950,843 - Filed: 2018-04-11 - Published: 2019-10-17 - Earliest claimed: 2018-04-11 - Assignees: Microsoft Technology Licensing, LLC - Applicants: Microsoft Technology Licensing, LLC - Inventors: Gal TAMIR; Rachel LEMBERG; Ze'ev MAOR; Ori BAR-ILAN - Web page: https://patentgrep.com/patent/US20190318010A1 ## CPC - G06F 40/166: Editing, e.g. inserting or deleting - G06F 3/167: {Audio in a user interface, e.g. using voice commands for navigating, audio feedback} - G06F 16/4393: {Multimedia presentations, e.g. slide shows, multimedia albums} - G06F 40/30: Semantic analysis - G10L 15/07: to the speaker - G10L 15/26: Speech to text systems (G10L15/08 takes precedence) - G06F 40/20: Natural language analysis (semantic analysis of natural language G06F40/30) - G10L 2015/228: {of application context} - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract Non-limiting examples of the present disclosure relate to automated presentation control during presentation of one or more digital documents. Processing operations described herein may be configured to work with any type of application/service, where examples described herein extend to presentation of any type of digital document(s) including slide-based presentations. Speech signals may be detected while a digital document is being presented. The detected speech signals may be converted to text. The converted text may be matched with contextual data (e.g., pre-processed content) associated with the digital document. Probability scoring for determination of a next placement position of a digital document may be generated based on the matching of converted text with the contextual data of the digital document. The next placement position may be selected based on the generated probability scoring. Display of the digital document may be automatically updated to reflect the selected next placement position. ## Claims ### Claim 1 A method comprising: presenting, in a user interface, a digital document; converting a speech signal to text; matching, using a language understanding model, the text with contextual data associated with the digital document; generating probability scoring for determination of a next placement position in the digital document by applying trained machine learning modeling that evaluates the text relative to the contextual data, wherein the applying of the trained machine learning modeling scores, for a plurality of portions of the digital document, a likelihood that the text is referring to a specific portion of the digital document; selecting, from scoring of the plurality of portions of the digital document, the specific portion of the digital document as the next placement position in the digital document based on the generated probability scoring; and automatically updating, through the user interface, a display of the digital document to the selected next placement position. ### Claim 2 (depends on 1) The method of claim 1, further comprising: detecting a current placement position of the digital document, and wherein the automatically updating comprises modifying display of the digital document from the current placement position to the selected next placement position. ### Claim 3 (canceled) ### Claim 4 (depends on 1) The method of claim 1, wherein the contextual data is one or more of: metadata associated with the digital document, one or more notes items associated with the digital document, and signal data from interaction with the user interface. ### Claim 5 (depends on 1) The method of claim 1, wherein the digital document is a slide-based presentation, wherein the contextual data is data associated with slides of the slide-based presentation, and wherein the selected next placement position is a different slide of the slide-based presentation. ### Claim 6 (depends on 5) The method of claim 5, wherein the automatically updating is a linear update to the digital document that changes, based on the selected next placement position, the display of the slide-based presentation to one of: a previous ordered slide of the slide-based presentation and a next ordered slide of the slide-based presentation. ### Claim 7 (depends on 5) The method of claim 5, wherein the automatically updating is a non-linear update to the digital document that changes, based on the selected next placement position, the display of the slide-based presentation to a slide that is out of a presentation order for the slide-based presentation. ### Claim 8 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: presenting, in a user interface, a digital document; converting a speech signal to text; matching, using a language understanding model, the text with contextual data associated with the digital document; generating probability scoring for determination of a next placement position in the digital document by applying trained machine learning modeling that evaluates the text relative to the contextual data, wherein the applying of the trained machine learning modeling scores, for a plurality of portions of the digital document, a likelihood that the text is referring to a specific portion of the digital document; selecting, from scoring of the plurality of portions of the digital document, the specific portion of the digital document as the next placement position in the digital document based on the generated probability scoring; and automatically updating, through the user interface, a display of the digital document to the selected next placement position. ### Claim 9 (depends on 8) The system of claim 8, where the method, executed by the at least one processor, further comprises: detecting a current placement position of the digital document, and wherein the automatically updating comprises modifying display of the digital document from the current placement position to the selected next placement position. ### Claim 10 (depends on 8) The system of claim 8, wherein the contextual data is trained practice speech from presentation of the digital document, and wherein the probability scoring is generated by comparing the text with the trained practice speech. ### Claim 11 (depends on 8) The system of claim 8, wherein the contextual data is one or more of: metadata associated with the digital document, one or more notes items associated with the digital document, and signal data from interaction with the user interface. ### Claim 12 (depends on 8) The system of claim 8, wherein the digital document is a slide-based presentation, wherein the contextual data is data associated with slides of the slide-based presentation, and wherein the selected next placement position is a different slide of the slide-based presentation. ### Claim 13 (depends on 12) The system of claim 12, wherein the automatically updating is a linear update to the digital document that changes, based on the selected next placement position, the display of the slide-based presentation to one of: a previous ordered slide of the slide-based presentation and a next ordered slide of the slide-based presentation. ### Claim 14 (depends on 12) The system of claim 12, wherein the automatically updating is a non-linear update to the digital presentation document that changes, based on the selected next placement position, the display of the slide-based presentation to a slide that is out of a presentation order for the slide-based presentation. ### Claim 15 A computer-readable storage medium storing computer-executable instructions that, when executed by at least one processor, causes the at least one processor to execute a method comprising: presenting, in a user interface, a digital document; converting a speech signal to text; matching, using a language understanding model, the text with contextual data associated with a digital document; generating probability scoring for determination of a next placement position in the digital document by applying trained machine learning modeling that evaluates the text relative to the contextual data, wherein the applying of the trained machine learning modeling scores, for a plurality of portions of the digital document, a likelihood that the text is referring to a specific portion of the digital document; selecting, from scoring of the plurality of portions of the digital document, the specific portion of the digital document as the next placement position in the digital document based on the generated probability scoring; and automatically updating, through the user interface, a display of the digital document to the selected next placement position. ### Claim 16 (depends on 15) The computer-readable storage medium of claim 15, where the executed method further comprising: detecting a current placement position of the digital document, and wherein the automatically updating comprises modifying display of the digital document from the current placement position to the selected next placement position. ### Claim 17 (depends on 15) The computer-readable storage medium of claim 15, wherein the contextual data is trained practice speech from presentation of the digital presentation document, and wherein the probability scoring is generated by comparing the text with the trained practice speech. ### Claim 18 (depends on 15) The computer-readable storage medium of claim 15, wherein the contextual data is one or more of: metadata associated with the digital document, one or more notes items associated with the digital document, and signal data from interaction with the user interface. ### Claim 19 (depends on 15) The computer-readable storage medium of claim 15, wherein the digital document is a slide-based presentation, wherein the contextual data is data associated with slides of the slide-based presentation, and wherein the selected next placement position is a different slide of the slide-based presentation. ### Claim 20 (depends on 19) The computer-readable storage medium of claim 19, wherein the automatically updating is a non-linear update to the digital document that changes, based on the selected next placement position, the display of the slide-based presentation to a slide that is out of a presentation order for the slide-based presentation. ### Claim 21 (depends on 1) The method of claim 1, further comprising: receiving, as the contextual data, presentation training data correlating previous user speech signals with specific portions of the digital document during a training run of a presentation of the digital document, and wherein the generating of the probability scoring further comprises scoring the plurality of portions of the digital document based on matching the text with the presentation training data correlating the previous user speech signals with specific portions of the digital document.