# US 10,504,523 B2 · Voice processing device, voice processing method, and computer program product > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,504,523 B2 - Document ID: US10504523B2 - Status: Granted - Kind: B2 - Application number: 15/891,075 - Filed: 2018-02-07 - Published: 2018-12-06 - Granted: 2019-12-10 - Earliest claimed: 2017-06-01 - Assignees: Kabushiki Kaisha Toshiba - Applicants: Kabushiki Kaisha Toshiba - Inventors: Ning Ding; Takehiko Kagoshima - Web page: https://patentgrep.com/patent/US10504523B2 ## CPC - 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 15/22: Procedures used during a speech recognition process, e.g. man-machine dialogue - H04L 12/1831: {Tracking arrangements for later retrieval, e.g. recording contents, participants activities or behavior, network status} - H04M 3/568: {audio processing specific to telephonic conferencing, e.g. spatial distribution, mixing of participants (echo suppression in two-way loud-speaking telephone systems H04M9/02; sound field processing per se H04S7/30)} - G10L 15/26: Speech to text systems (G10L15/08 takes precedence) - G10L 17/00: Speaker identification or verification techniques - G10L 21/0272: Voice signal separating - 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) - H04M: TELEPHONIC COMMUNICATION (circuits for controlling other apparatus via a telephone cable and not involving telephone switching apparatus G08) ## Abstract According to an embodiment, a voice processing device includes a receiver, a separator, and an output controller. The receiver is configured to receive n input signals input into n voice input devices respectively corresponding to n sound sources, where n is an integer of 2 or more. The separator is configured to separate the input signals by the sound sources to produce n separation signals. The output controller is configured to, according to the number of sound sources having uttered voice sounds, switch between an output signal produced based on the input signal and an output signal produced based on the separation signal, and output the output signal. ## Claims ### Claim 1 A voice processing device, comprising: a receiver configured to receive n input signals that is input into n voice input devices corresponding respectively to n sound sources, n being an integer of 2 or more; a separator configured to separate the input signals based at least in part on the sound sources to produce n separation signals; and an output controller configured to, according to a number of the sound sources that produce uttered voice sounds, switch between a first output signal produced based at least in part on the input signal and a second output signal produced based at least in part on the separation signal, wherein the output controller outputs the output signal. ### Claim 2 (depends on 1) The device according to claim 1, wherein, when the number of the sound sources that produce the uttered voice sounds is one, the output controller outputs the first output signal produced based at least in part on the input signal, and, when the number of the sound sources that produce the uttered voice sounds is two or more, the output controller outputs the second output signal produced based at least in part on the separation signal. ### Claim 3 (depends on 2) The device according to claim 2, wherein, when the number of the sound sources that produce the uttered voice sounds is one, the output controller produces the output signal emphasizing voice from a sound source direction, by beam-forming based at least in part on the input signal. ### Claim 4 (depends on 2) The device according to claim 2, wherein, when the number of the sound sources that produce the uttered voice sounds is one, the output controller selects one input signal from the n input signals, as the output signal. ### Claim 5 (depends on 1) The device according to claim 1, further comprising a recognition unit configured to perform voice recognition for the output signal. ### Claim 6 (depends on 1) The device according to claim 1, further comprising a determination unit configured to calculate a likelihood that the separation signal is a voice signal, and determine the number of the separation signals comprising the likelihood equal to or more than a threshold as being the number of the sound sources that produce the uttered voice sounds. ### Claim 7 (depends on 1) The device according to claim 1, further comprising a determination unit configured to determine the number of the separation signals comprising a power equal to or more than a threshold as being the number of the sound sources that produce the uttered voice sounds. ### Claim 8 (depends on 7) The device according to claim 7, wherein the threshold is an average of powers of the separation signals. ### Claim 9 A voice processing method, comprising: receiving n input signals acquired by n voice input devices respectively corresponding respectively to n sound sources, n being an integer of 2 or more; separating the input signals based at least in part on the sound sources to produce n separation signals; according to a number of the sound sources that produces uttered voice sounds, switching between a first output signal produced based at least in part on the input signal and a second output signal produced based at least in part on the separation signal; and outputting the output signal. ### Claim 10 A computer program product comprising a non-transitory computer readable medium including programmed instructions, the instructions, when executed by a computer, causing the computer to function as: a receiver configured to receive n input signals acquired by n voice input devices corresponding respectively to n sound sources, n being an integer of 2 or more; a separator configured to separate the input signals based at least in part on the sound sources to produce n separation signals; and an output controller configured to, according to a number of the sound sources that produce uttered voice sounds, switch between a first output signal produced based at least in part on the input signal and a second output signal produced based on the separation signal, wherein the output controller outputs the output signal.