# US 12,192,669 B2 · Auto focus on speaker during multi-participant communication conferencing > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,192,669 B2 - Document ID: US12192669B2 - Status: Granted - Kind: B2 - Application number: 17/835,120 - Filed: 2022-06-08 - Published: 2023-12-14 - Granted: 2025-01-07 - Earliest claimed: 2022-06-08 - Assignees: Avaya Management L.P. - Applicants: Avaya Management L.P. - Inventors: Prateek Singh; Divyarajsinh Jadeja; Mittali Jangid; Dean Beightol - Web page: https://patentgrep.com/patent/US12192669B2 ## CPC - H04N 5/2628: {Alteration of picture size, shape, position or orientation, e.g. zooming, rotation, rolling, perspective, translation} - G06V 20/40: in video content (extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44) - G06V 40/16: Human faces, e.g. facial parts, sketches or expressions - G06V 40/161: {Detection; Localisation; Normalisation} - G06V 40/172: {Classification, e.g. identification} - G06V 40/20: Movements or behaviour, e.g. gesture recognition (recognition of facial expressions G06V40/16) - G10L 17/06: Decision making techniques; Pattern matching strategies - G10L 25/57: for processing of video signals - H04L 65/403: Arrangements for multi-party communication, e.g. for conferences (data switching systems for conference H04L12/18; arrangements for connecting several subscribers to a common circuit, i.e. affording conference facilities H04M3/56; television conferencing systems H04N7/15) - G10L 17/00: Speaker identification or verification techniques - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING - H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M) ## Abstract A method for auto focus on a speaker during a communication session includes receiving video captured of a scene that includes a plurality of images of participants to the communication session, identifying the plurality of images of the participants in the video captured of the scene, recognizing audio from at least one of the participants, detecting facial movement in one of the images of the plurality of images and equating the recognized audio to the detected movement in the one of the images of the plurality of images. The method also includes selecting the one of the images of the plurality of images as a speaker based on the equated recognized audio to the detected movement in the one of the images, zooming in on the speaker and filtering out a remainder of the images of the plurality of images. ## Claims ### Claim 1 A method, comprising: receiving video captured of a scene that includes a plurality of images of participants to a communication session; identifying the plurality of images of the participants in the video captured of the scene based on registered facial prints; recognizing audio from at least one of the participants to the communication session based on registered voice prints; detecting facial movement in one of the images of the plurality of images using artificial intelligence; equating the recognized audio to the detected facial movement in the one of the images of the plurality of images using artificial intelligence; selecting the one of the images of the plurality of images as a speaker based on the equated recognized audio to the detected facial movement in the one of the images of the plurality of images and the registered facial prints and voice prints; zooming in on the speaker; and filtering out a remainder of the images of the plurality of images. ### Claim 2 (depends on 1) The method according to claim 1, wherein the communication session is a conference call. ### Claim 3 (depends on 1) The method according to claim 1, wherein identifying the plurality of images of the participants includes performing image recognition on the video to identify the plurality of images of the participants. ### Claim 4 (depends on 1) The method according to claim 1, wherein recognizing the audio from at least one of the participants to the communication session includes performing voice recognition to recognize the audio from at least one of the images. ### Claim 5 (depends on 1) The method according to claim 1, wherein detecting the facial movement in the one of the images of the plurality of images includes performing image recognition to detect the facial movement in the one of the images of the plurality of images. ### Claim 6 (depends on 1) The method according to claim 1, further comprising filtering out audio from other the participants to the communication session after zooming in on the speaker. ### Claim 7 (depends on 1) The method according to claim 1, further comprising zooming out on the speaker after not recognizing any audio from at least one of the participants to the communication session. ### Claim 8 (depends on 1) The method according to claim 1, further comprising selecting two of the images of the plurality of images as two speakers based on the equated recognized audio to the detected facial movement in the two of the images of the plurality of images and the registered facial prints and voice prints and zooming in on the two speakers. ### Claim 9 A system, comprising: a processor; and a memory coupled with and readable by the processor and having stored therein a set of instructions which, when executed by the processor, causes the processor to: receive video captured of a scene that includes a plurality of images of participants to a communication session; identify the plurality of images of the participants in the video captured of the scene based on registered facial prints; recognize audio from at least one of the participants to the communication session based on registered voice prints; detect facial movement in one of the images of the plurality of images using artificial intelligence; equate the recognized audio to the detected facial movement in the one of the images of the plurality of images using artificial intelligence; select the one of the images of the plurality of images as a speaker based on the equated recognized audio to the detected facial movement in the one of the images of the plurality of images and the registered facial prints and voice prints; zoom in on the speaker; and filter out a remainder of the images of the plurality of images. ### Claim 10 (depends on 9) The system according to claim 9, wherein the communication session is a conference call. ### Claim 11 (depends on 9) The system according to claim 9, wherein the instructions, when executed by the processor, further cause the processor to perform image recognition on the video to identify the plurality of images of the participants. ### Claim 12 (depends on 9) The system according to claim 9, wherein the instructions, when executed by the processor, further cause the processor to perform voice recognition to recognize the audio from at least one of the participants. ### Claim 13 (depends on 9) The system according to claim 9, wherein the instructions, when executed by the processor, further cause the processor to filter out audio from other participants to the communication session after zooming in on the speaker. ### Claim 14 (depends on 9) The system according to claim 9, wherein the instructions, when executed by the processor, further cause the processor to zoom out on the speaker after not recognizing any audio from at least one of the participants to the communication session. ### Claim 15 (depends on 9) The system according to claim 9, wherein the instructions, when executed by processor, further cause the processor to select two of the images of the plurality of images as two speakers based on the equated recognized audio to the detected facial movement in the two of the images of the plurality of images and the registered facial prints and voice prints and zoom in on the two speakers. ### Claim 16 A non-transitory computer readable medium comprising microprocessor executable instructions that, when executed by the microprocessor, perform the following functions: receive video captured of a scene that includes a plurality of images of participants to a communication session; identify the plurality of images of the participants in the video captured of the scene based on registered facial prints; recognize audio from at least one of the participants to the communication session based on registered voice prints; detect facial movement in one of the images of the plurality of images using artificial intelligence; equate the recognized audio to the detected facial movement in the one of the images of the plurality of images using artificial intelligence; select the one of the images of the plurality of images as a speaker based on the equated recognized audio to the detected facial movement in the one of the images of the plurality of images and the registered facial prints and voice prints; zoom in on the speaker; and filter out a remainder of the images of the plurality of images. ### Claim 17 (depends on 16) The non-transitory computer readable medium according to claim 16, wherein the communication session is a conference call. ### Claim 18 (depends on 16) The non-transitory computer readable medium according to claim 16, wherein the instructions, when executed by the microprocessor, further perform the function of perform image recognition on the video to identify the plurality of images of the participants. ### Claim 19 (depends on 16) The non-transitory computer readable medium according to claim 16, wherein instructions, when executed by the microprocessor, further perform the function of filter out audio from other participants to the communication session after zooming in on the speaker. ### Claim 20 (depends on 16) The non-transitory computer readable medium according to claim 16, wherein instructions, when executed by the microprocessor, further perform the function of select two of the images of the plurality of images as two speakers based on the equated recognized audio to the detected facial movement in the two of the images of the plurality of images and the registered facial prints and voice prints and zoom in on the two speakers.