IP Library Granted Patent US 12,566,792
Granted Patent B2
US 12,566,792 · App. 18/837,171 · Granted Mar 3, 2026

Systems and methods for automatically generating identifiers for multimedia information sets

Inventors: Lelai Deng (Beijing, CN); Junyu Dai (Beijing, CN); Xuchen Song (Los Angeles, CA); Shizhu Liu (Los Angeles, CA); Yuanjian Feng (Los Angeles, CA); Yufan Xue (Beijing, CN); Xiaohui Lin (Beijing, CN); Ce Gao (Beijing, CN)
Assignee: Lemon Inc.
G06F16/48G06F16/639
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,566,792
App. No.
18/837,171
Filed
Aug 8, 2024
Granted
Mar 3, 2026
Kind
B2
Art Unit
2165
USPC
707/736
Abstract

Embodiments of the present disclosure provide a method, apparatus, device, storage medium, and program for information processing. The method includes: obtaining a first multimedia information set that is to be processed and includes a plurality of first multimedia information; obtaining reference information that includes a plurality of reference multimedia identifiers, one of the reference multimedia identifiers indicating a reference multimedia information set that includes a plurality of reference multimedia information; determining a multimedia identifier to be selected according to the first multimedia information set, where a matching degree between a multimedia feature corresponding to a reference multimedia information set indicated by the multimedia identifier to be selected and a multimedia feature corresponding to the first multimedia information set satisfies a predetermined condition; and determining, according to the multimedia identifier to be selected, a target multimedia identifier corresponding to the first multimedia information set.

Claims (80)

1 . A method of automatically generating identifiers for multimedia information sets, comprising:

receiving a request to automatically generate an identifier for a first multimedia information set from a user device, wherein the first multimedia information set comprises a plurality of first multimedia items;

extracting features from the first multimedia information set based on receiving the request and classifying the extracted features in a plurality of dimensions to generate a first set of feature labels corresponding to the first multimedia information set;

extracting features from each of a plurality of reference multimedia information sets stored in a library and classifying the extracted features in the plurality of dimensions to generate a plurality of sets of reference feature labels, wherein each of the plurality of reference multimedia information sets corresponds to one of a plurality of reference multimedia identifiers, wherein each of the plurality of sets of reference feature labels corresponds to one of the plurality of reference multimedia identifiers, and wherein each of the plurality of reference multimedia information sets comprises a plurality of reference multimedia items;

computing a matching degree between the first set of feature labels and each of the plurality of sets of reference feature labels;

selecting at least one multimedia identifier from the plurality of reference multimedia identifiers based on determining that the matching degree between the first set of feature labels and the set of reference feature labels corresponding to the selected at least one multimedia identifier satisfies a predetermined condition; and

segmenting the selected at least one multimedia identifier into a plurality of terms and combining at least a portion of the plurality of terms to determine the identifier for the first multimedia information set.

2 . The method of claim 1 , wherein selecting the at least one multimedia identifier from the plurality of reference multimedia identifiers comprises:

determining a plurality of reference feature labels, according to the plurality of sets of reference feature labels, a repeated feature label being absent in the plurality of reference feature labels;

determining, according to the respective reference multimedia identifiers and the plurality of sets of reference feature labels, a set of reference multimedia identifiers corresponding to each of the reference feature labels; and

determining the at least one multimedia identifier, from the plurality of reference multimedia identifiers according to the first set of feature labels and the set of reference multimedia identifiers corresponding to each of the reference feature labels.

3 . The method of claim 2 , wherein selecting the at least one multimedia identifier from the plurality of reference multimedia identifiers comprises:

determining at least one target feature label from the plurality of reference feature labels according to the first set of feature labels, the target feature label belonging to the first set of feature labels; and

determining the at least one multimedia identifier to be selected, according to the set of reference multimedia identifiers corresponding to the at least one target feature label.

4 . The method of claim 1 , wherein combining at least the portion of the plurality of terms comprises:

combining the plurality of terms corresponding to the at least one multimedia identifier to be selected to obtain a plurality of recombined multimedia identifiers, the method further comprising:

modifying the plurality of recombined multimedia identifiers to obtain a plurality of modified multimedia identifiers; and

determining the target multimedia identifier from the plurality of modified multimedia identifiers.

5 . The method of claim 4 , wherein modifying the recombined multimedia identifiers to obtain the modified multimedia identifiers comprises for any one of the recombined multimedia identifiers:

performing at least one of the following on the recombined multimedia identifier to obtain the modified multimedia identifiers:

deleting a first predetermined term in the recombined multimedia identifiers;

replacing a predetermined type of terms in the recombined multimedia identifier, the predetermined type including a person name and/or a place name;

updating a term incompliant to a grammar in the recombined multimedia identifier;

performing synonym replacement on a term in the recombined multimedia identifier; and

deleting a recombined multimedia identifier with a number of characters less than or equal to a predetermined threshold, in the recombined multimedia identifier.

6 . The method of claim 4 , wherein determining the target multimedia identifier from the plurality of modified multimedia identifiers comprises:

determining a target reference multimedia identifier from the plurality of reference multimedia identifiers, a matching degree between a multimedia feature corresponding to the target reference multimedia identifier and the multimedia feature corresponding to the first multimedia information set being highest;

obtaining a similarity degree between each of the modified multimedia identifiers and the target reference multimedia identifier; and

determining the target multimedia identifier from the plurality of modified multimedia identifiers according to similarity degrees between respective modified multimedia identifiers of the modified multimedia identifiers and the target reference multimedia identifier.

7 . The method of claim 6 , wherein determining the target multimedia identifier from the plurality of modified multimedia identifiers according to the similarity degrees between the respective modified multimedia identifiers and the target reference multimedia identifier comprises:

sorting the plurality of modified multimedia identifiers, according to a descending order of the similarity degrees between the respective modified multimedia identifiers and the target reference multimedia identifier; and

determining first M modified multimedia identifiers after the sorting, as target multimedia identifiers, M being an integer greater than or equal to 1.

8 . The method of claim 1 , wherein each of the plurality of first multimedia items is any one of: an audio, a text, an image, or a video.

9 . The method of claim 1 , wherein each of the plurality of first multimedia items is a song, and where the plurality of dimensions comprises one or more of a style, an emotion, and a type.

10 . The method of claim 1 , after determining the identifier for the first multimedia information set, further comprising:

setting the identifier to be an identifier of the first multimedia information set; or

displaying the identifier; or

sending the identifier to a predetermined device.

11 . An electronic device, comprising:

a processor and a memory;

the memory storing computer executable instructions; and

the processor executing the computer executable instructions to implement acts comprising:

receiving a request to automatically generate an identifier for a first multimedia information set from a user device, wherein the first multimedia information set comprises a plurality of first multimedia items;

extracting features from the first multimedia information set based on receiving the request and classifying the extracted features in a plurality of dimensions to generate a first set of feature labels corresponding to the first multimedia information set;

extracting features from each of a plurality of reference multimedia information sets stored in a library and classifying the extracted features in the plurality of dimensions to generate a plurality of sets of reference feature labels, wherein each of the plurality of reference multimedia information sets corresponds to one of a plurality of reference multimedia identifiers, wherein each of the plurality of sets of reference feature labels corresponds to one of the plurality of reference multimedia identifiers, and wherein each of the plurality of reference multimedia information sets comprises a plurality of reference multimedia items;

computing a matching degree between the first set of feature labels and each of the plurality of sets of reference feature labels;

selecting at least one multimedia identifier from the plurality of reference multimedia identifiers based on determining that the matching degree between the first set of feature labels and the set of reference feature labels corresponding to the selected at least one multimedia identifier satisfies a predetermined condition; and

segmenting the selected at least one multimedia identifier into a plurality of terms and combining at least a portion of the plurality of terms to determine the identifier for the first multimedia information set.

12 . A non-transitory computer readable storage medium storing computer executable instructions, the computer executable instructions, when executed by a processor, implementing acts comprising:

receiving a request to automatically generate an identifier for a first multimedia information set from a user device, wherein the first multimedia information set comprises a plurality of first multimedia items;

extracting features from the first multimedia information set based on receiving the request and classifying the extracted features in a plurality of dimensions to generate a first set of feature labels corresponding to the first multimedia information set;

extracting features from each of a plurality of reference multimedia information sets stored in a library and classifying the extracted features in the plurality of dimensions to generate a plurality of sets of reference feature labels, wherein each of the plurality of reference multimedia information sets corresponds to one of a plurality of reference multimedia identifiers, wherein each of the plurality of sets of reference feature labels corresponds to one of the plurality of reference multimedia identifiers, and wherein each of the plurality of reference multimedia information sets comprises a plurality of reference multimedia items;

computing a matching degree between the first set of feature labels and each of the plurality of sets of reference feature labels;

selecting at least one multimedia identifier from the plurality of reference multimedia identifiers based on determining that the matching degree between the first set of feature labels and the set of reference feature labels corresponding to the selected at least one multimedia identifier satisfies a predetermined condition; and

segmenting the selected at least one multimedia identifier into a plurality of terms and combining at least a portion of the plurality of terms to determine the identifier for the first multimedia information set.

13 . The electronic device of claim 11 , wherein selecting the at least one multimedia identifier from the plurality of reference multimedia identifiers comprises:

determining a plurality of reference feature labels, according to the plurality of sets of reference feature labels, a repeated feature label being absent in the plurality of reference feature labels;

determining, according to the respective reference multimedia identifiers and the plurality of sets of reference feature labels, a set of reference multimedia identifiers corresponding to each of the reference feature labels; and

determining the at least one multimedia identifier, from the plurality of reference multimedia identifiers according to the first set of feature labels and the set of reference multimedia identifiers corresponding to each of the reference feature labels.

14 . The electronic device of claim 13 , wherein selecting the at least one multimedia identifier from the plurality of reference multimedia identifiers comprises:

determining at least one target feature label from the plurality of reference feature labels according to the first set of feature labels, the target feature label belonging to the first set of feature labels; and

determining the at least one multimedia identifier to be selected, according to the set of reference multimedia identifiers corresponding to the at least one target feature label.

15 . The electronic device of claim 11 , wherein combining at least the portion of the plurality of terms comprises:

combining the plurality of terms corresponding to the at least one multimedia identifier to be selected to obtain a plurality of recombined multimedia identifiers, the method further comprising:

modifying the plurality of recombined multimedia identifiers to obtain a plurality of modified multimedia identifiers; and

determining the target multimedia identifier from the plurality of modified multimedia identifiers.

16 . The electronic device of claim 15 , wherein modifying the recombined multimedia identifiers to obtain the modified multimedia identifiers comprises for any one of the recombined multimedia identifiers:

performing at least one of the following on the recombined multimedia identifier to obtain the modified multimedia identifiers;

deleting a first predetermined term in the recombined multimedia identifiers;

replacing a predetermined type of terms in the recombined multimedia identifier, the predetermined type including a person name and/or a place name;

updating a term incompliant to a grammar in the recombined multimedia identifier;

performing synonym replacement on a term in the recombined multimedia identifier; and

deleting a recombined multimedia identifier with a number of characters less than or equal to a predetermined threshold, in the recombined multimedia identifier.

17 . The electronic device of claim 15 , wherein determining the target multimedia identifier from the plurality of modified multimedia identifiers comprises:

determining a target reference multimedia identifier from the plurality of reference multimedia identifiers, a matching degree between a multimedia feature corresponding to the target reference multimedia identifier and the multimedia feature corresponding to the first multimedia information set being highest;

obtaining a similarity degree between each of the modified multimedia identifiers and the target reference multimedia identifier; and

determining the target multimedia identifier from the plurality of modified multimedia identifiers according to similarity degrees between respective modified multimedia identifiers of the modified multimedia identifiers and the target reference multimedia identifier.

18 . The electronic device of claim 17 , wherein determining the target multimedia identifier from the plurality of modified multimedia identifiers according to the similarity degrees between the respective modified multimedia identifiers and the target reference multimedia identifier comprises:

sorting the plurality of modified multimedia identifiers, according to a descending order of the similarity degrees between the respective modified multimedia identifiers and the target reference multimedia identifier; and

determining first M modified multimedia identifiers after the sorting, as target multimedia identifiers, M being an integer greater than or equal to 1.

Assignments (8)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 2, 2026
From: LIN, XIAOHUI
To: TIKTOK INFORMATION TECHNOLOGIES UK LIMITED
Reel/Frame 073934/0889 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 4, 2026
From: DAI, JUNYU; XUE, YUFAN; GAO, CE
To: SHANGHAI SUIXUNTONG ELECTRONIC TECHNOLOGY CO., LTD.
Reel/Frame 073685/0655 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 4, 2026
From: MIAOZHENDIDA (BEIJING) NETWORK TECHNOLOGY CO., LTD.
To: LEMON INC.
Reel/Frame 073685/0865 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 4, 2026
From: SHANGHAI SUIXUNTONG ELECTRONIC TECHNOLOGY CO., LTD.
To: LEMON INC.
Reel/Frame 073686/0037 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 4, 2026
From: BYTEDANCE INC.
To: LEMON INC.
Reel/Frame 073686/0119 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 4, 2026
From: TIKTOK INFORMATION TECHNOLOGIES UK LIMITED
To: LEMON INC.
Reel/Frame 073686/0165 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 4, 2026
From: LIU, SHIZHU
To: BYTEDANCE INC.
Reel/Frame 073685/0592 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 4, 2026
From: DENG, LELAI
To: MIAOZHENDIDA (BEIJING) NETWORK TECHNOLOGY CO., LTD.
Reel/Frame 074605/0010 →
Priority Claims (1)
CN 202210154951.9 · Feb 21, 2022 · national
Continuity (1)
Related Publication 20250173371A1 · May 29, 2025
References Cited (58)
US 7035467B2 · Nicponski · 2006 [cited by examiner]
US 8316041B1 · Chang · 2012 [cited by examiner]
US 8953228B1 · Mehers · 2015 [cited by examiner]
US 10657098B2 · Allen · 2020 [cited by examiner]
US 10957290B2 · Nazer · 2021 [cited by examiner]
US 20020159640A1 · Vaithilingam · 2002 [cited by examiner]
US 20030128877A1 · Nicponski · 2003 [cited by examiner]
US 20040078383A1 · Mercer et al. · 2004 [cited by applicant]
US 20060230331A1 · Abanami · 2006 [cited by examiner]
US 20070055697A1 · Luo · 2007 [cited by examiner]
US 20070067157A1 · Kaku · 2007 [cited by examiner]
US 20090150445A1 · Herberger et al. · 2009 [cited by applicant]
US 20100057694A1 · Kunjithapatham · 2010 [cited by examiner]
US 20100082615A1 · Clinchant · 2010 [cited by examiner]
US 20100226582A1 · Luo · 2010 [cited by examiner]
US 20100278396A1 · Mitsuhashi · 2010 [cited by examiner]
US 20110314367A1 · Chang · 2011 [cited by examiner]
US 20120232681A1 · Mundy · 2012 [cited by examiner]
US 20120269436A1 · Mensink · 2012 [cited by examiner]
US 20120281887A1 · Yamaguchi · 2012 [cited by examiner]
US 20120296908A1 · Bach · 2012 [cited by examiner]
US 20140328570A1 · Cheng · 2014 [cited by examiner]
US 20150055854A1 · Marchesotti · 2015 [cited by examiner]
US 20150169542A1 · Lin · 2015 [cited by examiner]
US 20150169952A1 · O'Malley · 2015 [cited by examiner]
US 20150310011A1 · Poulet · 2015 [cited by examiner]
US 20160328096A1 · Tran · 2016 [cited by examiner]
US 20160343088A1 · Bhasin · 2016 [cited by examiner]
US 20170329697A1 · Dillinger · 2017 [cited by examiner]
US 20170330540A1 · Quattro · 2017 [cited by examiner]
US 20180124437A1 · Memisevic · 2018 [cited by examiner]
US 20190332867A1 · Zhang · 2019 [cited by examiner]
US 20190354578A1 · Shen · 2019 [cited by examiner]
US 20200151585A1 · Ishida · 2020 [cited by examiner]
US 20200336802A1 · Russell · 2020 [cited by examiner]
US 20200364338A1 · Ducau · 2020 [cited by examiner]
US 20200372066A1 · Saggi · 2020 [cited by examiner]
US 20210034657A1 · Kale · 2021 [cited by examiner]
US 20210124997A1 · Kondou · 2021 [cited by examiner]
US 20210271707A1 · Lin · 2021 [cited by examiner]
US 20220067576A1 · Saha · 2022 [cited by examiner]
US 20220237373A1 · Singh Bawa · 2022 [cited by examiner]
US 20220292685A1 · Heisler · 2022 [cited by examiner]
US 20220350838A1 · Henderson · 2022 [cited by examiner]
US 20230402065A1 · Gu · 2023 [cited by examiner]
US 20230409633A1 · Lebecque · 2023 [cited by examiner]
CN 107193878A · 2017 [cited by applicant]
CN 107463552A · 2017 [cited by applicant]
CN 107193878B · 2020 [cited by applicant]
CN 111125387A · 2020 [cited by applicant]
CN 112487236A · 2021 [cited by applicant]
CN 110096571B · 2021 [cited by applicant]
CN 113282788A · 2021 [cited by applicant]
CN 113569064A · 2021 [cited by applicant]
Kim, Haven, Seungheon Doh, Junwon Lee, and Juhan Nam. “Music playlist title generation using artist information.” (2023) rXiv preprint arXiv:2301.08145 (Year: 2023). [cited by examiner]
Yun Hao, “Towards a Better Understanding of Music Playlist Titles and Descriptions”, 2021, University of Illinois Urbana-Champaign (Year: 2021). [cited by examiner]
International Patent Application No. PCT/SG2023/050098; Int'l Search Report; dated Sep. 20, 2023; 3 pages. [cited by applicant]
European Patent Application No. 23756738.3; Extended European Search Report dated Feb. 28, 2025, 8 pages. [cited by applicant]