# US 9,520,131 B2 · Apparatus and method for voice processing > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 9,520,131 B2 - Document ID: US9520131B2 - Status: Granted - Kind: B2 - Application number: 14/146,015 - Filed: 2014-01-02 - Published: 2014-10-16 - Granted: 2016-12-13 - Earliest claimed: 2013-04-11 - Assignees: Wistron Corporation - Applicants: Wistron Corporation - Inventors: Hsi-Chun Hsiao; Shou-Te Wei - Web page: https://patentgrep.com/patent/US9520131B2 ## CPC - G10L 15/25: using position of the lips, movement of the lips or face analysis - G10L 25/78: Detection of presence or absence of voice signals (switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10) - G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING ## Abstract An apparatus and a corresponding method for voice processing are provided. The apparatus includes a sound receiver, a camera, and a processor. The sound receiver receives a sound signal. The camera takes a video. The processor is coupled to the sound receiver and the camera. The processor obtains a voice onset time (VOT) of the sound signal, detects a human face in the video, detects a change time of a mouth contour of the human face, and verifies at least one preset condition. When all of the preset conditions are true, the processor performs speech recognition on the sound signal. The at least one preset condition includes that a difference between the VOT and the change time is smaller than a threshold value. ## Claims ### Claim 1 A voice processing apparatus, comprising: a sound receiver, receiving a sound signal; a camera, taking a video; and a processor, coupled to the sound receiver and the camera, obtaining a voice onset time of the sound signal, detecting a human face in the video, detecting a change time of a mouth contour of the human face, after detecting a change amount of the mouth contour between two frames of the video, and setting the change time as a time when the change amount is greater than a second threshold value, verifying at least one preset condition, and performing speech recognition on the sound signal when the at least one preset condition is all true, wherein the at least one preset condition comprises that a difference between the voice onset time and the change time is smaller than a first threshold value, wherein the processor determines a sound source direction of the sound signal, and the at least one preset condition comprises that an angle difference between the sound source direction and a preset direction is smaller than a third threshold value. ### Claim 2 (depends on 1) The voice processing apparatus as claimed in claim 1, wherein the change amount is a change amount in area, height or width of the mouth contour. ### Claim 3 (depends on 1) The voice processing apparatus as claimed in claim 1, wherein the processor detects a circumscribing quadrangle of the mouth contour, and the change amount is a change amount in area, height or width of the circumscribing quadrangle. ### Claim 4 (depends on 1) The voice processing apparatus as claimed in claim 1, wherein the sound receiver comprises a first microphone and a second microphone, the sound signal comprises a first sound signal and a second sound signal, the first microphone and the second microphone respectively receive the first sound signal and the second sound signal, and the processor determines the sound source direction according to a time difference or a phase difference between the first sound signal and the second sound signal. ### Claim 5 (depends on 1) The voice processing apparatus as claimed in claim 1, wherein the processor determines a sound source direction of the sound signal, and converts a position of the human face in a frame of the video into a direction of the human face according to a look-up table, and the at least one preset condition comprises that an angle difference between the sound source direction and the direction of the human face is smaller than a fourth threshold value. ### Claim 6 (depends on 5) The voice processing apparatus as claimed in claim 5, wherein the processor records a plurality of positions of the human face in frames of the video and the sound source direction corresponding to each of the positions, so as to establish the look-up table. ### Claim 7 A method for voice processing, comprising: receiving a sound signal; obtaining a voice onset time of a sound signal; taking a video; detecting a human face in the video; detecting a change time of a mouth contour of the human face; detecting a change amount of the mouth contour between two frames of the video; setting the change time as a time when the change amount is greater than a second threshold value; verifying at least one preset condition; performing speech recognition on the sound signal when the at least one preset condition is all true, wherein the at least one preset condition comprises that a difference between the voice onset time and the change time is smaller than a first threshold value; and determining a sound source direction of the sound signal, wherein the at least one preset condition comprises that an angle difference between the sound source direction and a preset direction is smaller than a third threshold value. ### Claim 8 (depends on 7) The method for voice processing as claimed in claim 7, wherein the change amount is a change amount in area, height or width of the mouth contour. ### Claim 9 (depends on 7) The method for voice processing as claimed in claim 7, wherein the step of detecting the change amount of the mouth contour comprises: detecting a circumscribing quadrangle of the mouth contour, wherein the change amount is a change amount in area, height or width of the circumscribing quadrangle. ### Claim 10 (depends on 7) The method for voice processing as claimed in claim 7, wherein the sound signal comprises a first sound signal and a second sound signal, and the step of determining the sound source direction comprises: determining the sound source direction according to a time difference or a phase difference between the first sound signal and the second sound signal. ### Claim 11 (depends on 7) The method for voice processing as claimed in claim 7, further comprising: determining a sound source direction of the sound signal; and converting a position of the human face in a frame of the video into a direction of the human face according to a look-up table, wherein the at least one preset condition comprises that an angle difference between the sound source direction and the direction of the human face is smaller than a fourth threshold value. ### Claim 12 (depends on 11) The method for voice processing as claimed in claim 11, further comprising: recording a plurality of positions of the human face in frames of the video, and the sound source direction corresponding to each of the positions, so as to establish the look-up table.