# US 2023/0410831 A1 · Electronic apparatus and controlling method thereof > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0410831 A1 - Document ID: US20230410831A1 - Status: Published application - Kind: A1 - Application number: 18/204,190 - Filed: 2023-05-31 - Published: 2023-12-21 - Earliest claimed: 2022-05-23 - Assignees: Samsung Electronics Co., Ltd. - Applicants: Samsung Electronics Co., Ltd. - Inventors: Soofeel KIM; Jiwon KIM; Jongtae KIM; Jaeyong HWANG - Web page: https://patentgrep.com/patent/US20230410831A1 ## CPC - G10L 21/028: using properties of sound source - G06N 20/00: Machine learning - G10L 17/02: Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction - G10L 17/00: Speaker identification or verification techniques - 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 - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract An electronic apparatus includes: a memory storing at least one instruction; and at least one processor configured to divide audio data into a plurality of periods to include overlapping regions, acquire an audio feature from each of the plurality of divided periods, identify a first audio source and a second audio source in each of the plurality of divided periods based on the audio feature, and acquire first audio data corresponding to the first audio source and second audio data corresponding to the second audio source from the audio data. ## Claims ### Claim 1 An electronic apparatus comprising: a memory configured to store at least one instruction; and at least one processor configured to execute the at least one instruction to: divide audio data into a plurality of sections of audio data based on a plurality of time periods having a plurality of overlapping regions, acquire an audio feature from each of the plurality of sections, identify a first audio source and a second audio source in each of the plurality of sections based on the audio feature, and acquire first audio data corresponding to the first audio source and second audio data corresponding to the second audio source from the audio data. ### Claim 2 (depends on 1) The electronic apparatus as claimed in claim 1, wherein the at least one processor is further configured to divide the audio data to: acquire a first section corresponding to a first period, among the plurality of time periods, and acquire a second corresponding to a second period, among the plurality of time periods, wherein the second period is different from the first period, and wherein the first section and the second section comprises an overlapping region, among the overlapping regions. ### Claim 3 (depends on 1) The electronic apparatus as claimed in claim 1, wherein the at least one processor is configured to: identify a plurality of audio blocks based on the first audio source or the second audio source in each of the plurality of time periods; acquire the first audio data by combining two or more first audio blocks, among the plurality of audio blocks, corresponding to the first audio source with each other; and acquire the second audio data by combining two or more second audio blocks, among the plurality of audio blocks, corresponding to the second audio source with each other. ### Claim 4 (depends on 3) The electronic apparatus as claimed in claim 3, wherein the at least one processor is further configured to: identify the overlapping regions in the plurality of audio blocks, and acquire the first audio data and the second audio data by comparing the plurality of overlapping regions with each other. ### Claim 5 (depends on 3) The electronic apparatus as claimed in claim 3, wherein the at least one processor is further configured to: acquire similarity between the plurality of overlapping regions in the plurality of audio blocks, acquire the first audio data by combining the two or more first audio blocks having similarity greater than or equal to a first threshold value; and acquire the second audio data by combining the two or more second audio blocks having similarity greater than or equal to the first threshold value. ### Claim 6 (depends on 5) The electronic apparatus as claimed in claim 5, wherein the at least one processor is further configured to: extend at least one of the overlapping regions when the similarity is less than the first threshold value and greater than or equal to a second threshold value, and divide the audio data based on the extended overlapping region. ### Claim 7 (depends on 3) The electronic apparatus as claimed in claim 3, wherein the at least one processor is further configured to: acquire a peak value of an audio signal in a first audio block among the plurality of audio blocks, acquire an average value of the audio signal included in an overlapping region of the first audio block, and identify whether the overlapping region is silent based on a difference between the peak value and the average value. ### Claim 8 (depends on 1) The electronic apparatus as claimed in claim 1, wherein the at least one processor is further configured to: identify a number of audio sources included in the audio data, and divide the audio data when the number of audio sources that is identified is less than a threshold number. ### Claim 9 (depends on 8) The electronic apparatus as claimed in claim 8, wherein the at least one processor is further configured to divide the audio data when a duration of the audio data is greater than or equal to a threshold time. ### Claim 10 (depends on 8) The electronic apparatus as claimed in claim 8, wherein the at least one processor is configured to: identify the number of audio sources in the audio data based on a first artificial intelligence model stored in the memory, and acquire the first audio data and the second audio data based on a second artificial intelligence model stored in the memory, the first artificial intelligence model being different from the first artificial intelligence model. ### Claim 11 A controlling method of an electronic apparatus, the method comprising: dividing audio data into a plurality of sections of audio data based on a plurality of time periods having a plurality of overlapping regions; acquiring an audio feature from each of the plurality of sections; identifying a first audio source and a second audio source in each of the plurality of sections based on the audio feature; and acquiring first audio data corresponding to the first audio source and second audio data corresponding to the second audio source from the audio data. ### Claim 12 (depends on 11) The method as claimed in claim 11, wherein the dividing the audio data comprises: acquiring a first section corresponding to a first period, among the plurality of time periods, and acquiring a second section corresponding to a second period, among the plurality of time periods, wherein the second period is different from the first period, and wherein the first section and the second section comprises an overlapping region, among the overlapping regions. ### Claim 13 (depends on 11) The method as claimed in claim 11, further comprising: identifying a plurality of audio blocks based on the first audio source or the second audio source in each of the plurality of time periods, wherein the acquiring the first audio data and the second audio data comprises: acquiring the first audio data by combining two or more first audio blocks, among the plurality of audio blocks, corresponding to the first audio source with each other; and acquiring the second audio data by combining two or more second audio blocks, among the plurality of audio blocks, corresponding to the second audio source with each other. ### Claim 14 (depends on 13) The method as claimed in claim 13, wherein the acquiring of the first audio data and the second audio data comprises: identifying the plurality of overlapping regions in the plurality of audio blocks, and acquiring the first audio data and the second audio data by comparing the plurality of overlapping regions with each other. ### Claim 15 (depends on 14) The method as claimed in claim 14, further comprising: acquiring similarity between the plurality of overlapping regions in the plurality of audio blocks, wherein the acquiring the first audio data and the second audio data comprises: acquiring the first audio data by combining the two or more first audio blocks having similarity greater than or equal to a first threshold value; and acquiring the second audio data by combining the two or more second audio blocks having similarity greater than or equal to the first threshold value.