CN105138670A - Audio file label generation method and system - Google Patents
Audio file label generation method and system Download PDFInfo
- Publication number
- CN105138670A CN105138670A CN201510564305.XA CN201510564305A CN105138670A CN 105138670 A CN105138670 A CN 105138670A CN 201510564305 A CN201510564305 A CN 201510564305A CN 105138670 A CN105138670 A CN 105138670A
- Authority
- CN
- China
- Prior art keywords
- audio file
- label
- word
- keyword
- storehouse
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to an audio file label generation method and system. The method comprises the steps that an audio file needing label setting and text information contained in content of the audio file are input into an audio file library, a common label is input into a label library, and a common key word is input into a key word library; according to the correlation degree of the common key word and the common label, the label is set for the common key word, and a common key word-label relation is obtained; the text information is segmented to obtain a plurality of words, the TF-IDF value of each word is calculated, and a key word of the audio file is obtained according to the TF-IDF values; according to the common key word-label relation, a label corresponding to the key word of the audio file is searched for; if the corresponding label is found, the audio file is associated with the corresponding label; a new label is generated for a new word, and the new word and the new label are associated, and the new label is set as the label of the audio file.
Description
Technical field
The present invention relates to application service technical field, particularly relate to a kind of audio file label generating method and system.
Background technology
Along with the development of network technology and masses are for the change of receives information mode, the audio file quantity on network is explosive growth.For digital music, by the end of the year 2014, China Internet music user will have reached 4.78 hundred million, daily download more than 200,000,000 times, and estimate between 2 to three years of future, and music netizen number of users is by breakthrough 600,000,000, and annual download will more than 100,000,000,000 times.Meanwhile, existing music is hundreds of millions of, and the current each digital music platform of main flow, the music volume of receipts is all more than 1,000,000.In the face of so huge customer volume and audio file amount, a key issue urgently to be resolved hurrily is exactly how to be displayed, to improve Consumer's Experience in batches by the audio file meeting user's needs more accurately.A kind of conventional mode is for audio file adds label.
Current audio file label generating platform generally comprises three kinds of modes:
1, human configuration label;
2, crawl internet label, namely adopt crawlers to be that the label that audio file configures is brought as oneself is used by other platforms.
3, user participates in, and namely user is according to the label judging to upload as audio file configures of oneself, and program screens most suitable label according to label statistics.
But the mode efficiency of human configuration label is low, and crawl label or user to upload the mode accuracy rate of label low from internet.
Summary of the invention
Based on this, be necessary low for prior art efficiency, that accuracy rate is low problem, a kind of audio file label generating method and system are provided.
A kind of audio file label generating method, comprises the following steps:
By the Word message inputting audio library needing to carry out comprising in the audio file of label setting and the content of audio file, label typing tag library will be commonly used, keyword typing keywords database will be commonly used;
According to the correlation degree of conventional keyword and conventional label, for conventional keyword arranges label, obtain conventional keyword-label relation; Wherein, described keyword-label relation represents the corresponding relation between keyword and corresponding label;
Participle is carried out to described Word message, obtains several words, calculate the TF-IDF value of each word, and according to the acquiring size of TF-IDF value the keyword of audio file; Wherein, described TF-IDF value represents the word frequency of word and the product of inverse document frequency;
The label corresponding according to the keyword of conventional keyword-label relational query audio file; If inquire corresponding label, audio file is associated with corresponding label;
For new word generates new label, and associate described new word and described new label, described new label is set to the label of audio file; Wherein, described new word is the keyword of the audio file not inquiring corresponding label.
Above-mentioned audio file label generating method, by carrying out participle to the Word message comprised in the content of audio file, calculate the size of the TF-IDF value of each word thus obtain the keyword of described audio file, the label corresponding according to the keyword of conventional keyword-label relational query audio file; If inquire corresponding label, audio file is associated with corresponding label; Otherwise the keyword of audio file is added new word storehouse, and the new word of making a living in dictionary generates new label, and associate described new word and described new label, described new label is set to the label of audio file, execution efficiency is high, and result accuracy is high.
A kind of audio file label creation system, comprising:
Label generation server, audio file show server, audio file data storehouse server, system management server;
Label generation server is audio file correlation tag according to the Word message comprised in the content of the audio file in the server of audio file data storehouse and audio file;
Audio file shows that server extracts the data of audio file from the server of audio file data storehouse according to user operation, and by described data display to client;
System management server provides audio file data management service.
Above-mentioned audio file label creation system is audio file correlation tag by label generation server according to the Word message comprised in the content of the audio file in the server of audio file data storehouse and audio file; From the server of audio file data storehouse, extracted the data of audio file according to user operation by audio file displaying server, and by described data display to client; And provide audio file data management service by system management server, provide a kind of intelligentized song label creation system, improve efficiency and the accuracy of label generation.
Accompanying drawing explanation
Fig. 1 is the audio file label generating method process flow diagram of an embodiment;
Fig. 2 is the TF-IDF value calculating method process flow diagram of an embodiment;
Fig. 3 is the structural representation of the audio file label creation system of an embodiment;
Fig. 4 is the structural representation of the audio file library of an embodiment.
Embodiment
Below in conjunction with accompanying drawing, technical scheme of the present invention is described further.
As shown in Figure 1, audio file label generating method of the present invention comprises the following steps:
S1, by the Word message inputting audio library needing to carry out comprising in the audio file of label setting and the content of audio file, will commonly use label typing tag library, will commonly use keyword typing keywords database;
S2, according to the correlation degree of conventional keyword and conventional label, for conventional keyword arranges label, obtains conventional keyword-label relation; Wherein, described keyword-label relation represents the corresponding relation between keyword and corresponding label;
S3, carries out participle to described Word message, obtains several words, calculates the TF-IDF value of each word, and according to the acquiring size of TF-IDF value the keyword of audio file; Wherein, described TF-IDF value represents the word frequency of word and the product of inverse document frequency;
S4, the label corresponding according to the keyword of conventional keyword-label relational query audio file; If inquire corresponding label, audio file is associated with corresponding label;
S5, for new word generates new label, and associates described new word and described new label, described new label is set to the label of audio file; Wherein, described new word is the keyword of the audio file not inquiring corresponding label.
In step sl, the Word message inputting audio library that can will comprise in the content of audio file and audio file, for typing song, can by the lyrics inputting audio library of song, also can according to actual needs by song title, issuing time, clicks, album name, listen to the inputting audio libraries such as user group.In addition, also can by conventional label typing tag library, keyword typing keywords database will be commonly used.
Wherein, described audio file library can comprise label generation server, audio file shows server, audio file data storehouse server, system management server; Wherein, label generation server can be audio file correlation tag according to the Word message comprised in the content of the audio file in the server of audio file data storehouse and audio file; Audio file shows that server can extract the data of audio file from the server of audio file data storehouse according to user operation, and by described data display to client; System management server can provide audio file data management service.
In step s 2, according to the correlation degree of conventional keyword and conventional label, for conventional keyword arranges label, conventional keyword-label relation can be obtained; Wherein, described keyword-label relation represents the corresponding relation between keyword and corresponding label.
Be still song for audio file, conventional label can comprise sings bent wind, languages etc.; Conventional keyword can comprise popular song, ethnic song, Japan and Korea S's song, American-European song etc.Correspondingly, during for above-mentioned keyword correlation tag, keyword popular song, ethnic song can be signed with performance Qu Fengbiao and associate, keyword Japan and Korea S song, American-European song are associated with languages label.
In step s3, can adopt segmenter to audio file content in the Word message that comprises carry out participle.Such as, jcseg segmenter is adopted.Jcseg segmenter provides following three kinds of cutting patterns:
(1) ((ForwardMaximumMatchingMethod, Forward Maximum Method) algorithm is applicable to rate request occasion to simple model: FMM.
(2) complex patterns: MMSEG tetra-kinds of filter algorithms, have higher qi justice and remove, participle accuracy rate can reach 98.41%, can match synonym simultaneously.
(3) detecting pattern: only return existing entry in dictionary, be well suited for some application scenario.
Different cutting patterns can be selected according to actual conditions.Such as, in order to label allocation more accurately, can complex patterns be adopted, can 479338 words/second, the participle speed of 1324.4KB/ second carries out participle.
In practical situations both, also can require to adopt other segmenter, carry out participle with other participle speed according to algorithm, complexity, accuracy.
In order to the demand that labels of the characteristic or audio file with exclusive copyright that embody different platform, when adopting segmenter to carry out participle, also can according to new terms such as platform characteristic word, programm name, network vocabulary, segmenter dictionary is expanded, then adopts the segmenter after expansion to carry out participle to the Word message comprised in the content of audio file.
The concrete implementing procedure of step S3 is as follows:
S31, is divided into several participles by the Word message comprised in the content of audio file;
S32, filters the stop words in participle and punctuation mark, obtains useful word;
S33, calculates word frequency and the inverse document frequency of each useful word;
S34, calculates the TF-IDF value of each word according to described word frequency and inverse document frequency;
S35, the keyword of audio file according to the acquiring size of described TF-IDF value.
Wherein, described word frequency (TermFrequency, TF) represent the frequency that in a song, certain word occurs, described inverse document frequency (termfrequency – inversedocumentfrequency, IDF) represents the weights of importance of each word.
The word frequency of each word and inverse document frequency are respectively according to following formulae discovery:
The account form of described inverse document frequency is:
tfidf
i,j＝tf
i,j×idf
i(3)
In formula, tf
i,jrepresent i-th the unduplicated useful word t obtained from a jth audio file
iword frequency; n
i,jrepresent the useful word t obtained from a jth audio file
iquantity; Σ
kn
k,jrepresent the occurrence number sum of all useful word obtained from a jth audio file, idf
irepresent word t
iinverse document frequency, | D| is the sum of the Word message in audio file library, | { j:t
i∈ d
j| for comprising word t in audio file library
ithe quantity of Word message, tfidf
i,jbe i-th unduplicated useful word t
itF-IDF value in a jth audio file.
For example, if total the word number of an audio file is 100, and word " cow " has occurred 3 times, and so the word frequency of " cow " word in this audio file is exactly 0.03 (3/100); " if cow " one word occurred in 1,000 audio file of audio file library, and the audio file sum in audio file library is 10,000,000, its reverse audio file frequency is exactly 4 (log (10,000,000/1,000)).Last TF-IDF value is 0.12 (0.03*4).
The TF-IDF value calculating method of other words is identical with said method, repeats no more herein.
The keyword of audio file can be selected according to the TF-IDF value of each word.Such as, the keyword of word as audio file of TF-IDF value arrangement front 3 can be extracted.According to actual needs, the keyword of word as audio file of TF-IDF value arrangement front 2 or front 4 can also be selected.
In one embodiment, can calculate according to under type | { j:t
i∈ d
j|:
From inverse document frequency storehouse, inquiry packet is containing go-word language t
iword message;
If do not inquired, then will include and use word t
ithe record of Word message be inserted in inverse document frequency storehouse, and to make | { j:t
i∈ d
j| be 1; Otherwise, will | { j:t
i∈ d
j| add 1.
In one embodiment, if do not inquire label corresponding to the keyword of audio file according to step S4, also following operation can be performed:
Whether containing described new word in inquiry new word storehouse;
If contained, then the quantity of new word described in new word storehouse is added 1; Otherwise, described new word is inserted in new word storehouse, and the quantity of this keyword is set to 1.
Such as, comprise " cow " word in the word segmentation result of audio file, and in conventional keyword-label relation, do not inquire label corresponding to " cow " word, then to inquire about in new word storehouse whether containing " cow " word.If containing " cow " word in new word storehouse, and " cow " one the quantity of word be 2, then by " cow " one the quantity of word add 1, obtain " cow " one the quantity of word be 3; If new word does not comprise " cow " word in storehouse, then by " cow " one word add in new word storehouse, and by " cow " one the quantity of word be set to 1.
In one embodiment, also can be audio file associated user group label.Concrete mode is as follows:
According to user characteristics, the user of audio file is divided into multiple customer group;
For audio file is set up and associated user group label, the initial value of described customer group label is set to sky; Wherein, described customer group label represents the customer group belonging to user of audio file;
According to the playing duration of audio file in different user group and broadcasting time, the attention rate of statistics audio file in different user group;
Customer group label the highest for attention rate is set to the customer group label of described audio file.
Described user characteristics comprises age, sex, occupation etc.Such as, according to age of user, customer group can be divided into " after 80s ", " after 90s "; Or be divided into the classification such as " teenager ", " person in middle and old age "; Also according to user's sex, customer group can be divided into " man ", " female ".
" after 80s ", " after 90s " is divided into, if the attention rate of certain audio file in " after 80s " set is the highest, then for this audio file inserts " after 80s " customer group label for customer group.
For ensureing the accuracy of the above results, a customer group analysis task can be performed at set intervals, such as, a customer group analysis task can be performed week about; Again such as, also a customer group analysis task can be performed every other month.
The attention rate of described audio file in different user group can be added up respectively according to under type:
P
i＝T
i/(T
0,i×T
1)，
In formula, P
irepresent the attention rate of audio file in i-th customer group, T
irepresent the total playing duration of audio file in i-th customer group, T
0, irepresent the broadcasting time of audio file in i-th customer group, T
1represent the duration of audio file.
When user plays described audio file, recording user is to the playing duration of described audio file, described playing duration is accumulated in the playing duration record of audio file described in respective user group, and the broadcasting time of audio file described in respective user group is added 1.
Above-mentioned audio file label generating method has the following advantages:
1) associate audio file label, saved manpower and time, efficiency is high;
2) according to new terms such as platform characteristic word, programm name, network vocabulary, segmenter dictionary can be expanded, the characteristic of different platform or the demand that labels of audio file with exclusive copyright can be embodied;
3) adopt unified standard to carry out label association, accuracy is high.
Be further described below in conjunction with the embodiment of accompanying drawing to audio file label creation system of the present invention.
As shown in Figure 3, audio file label creation system of the present invention comprises:
Label generation server 10, audio file show server 20, audio file data storehouse server 30, system management server 40;
Label generation server 10 is audio file correlation tag according to the data in audio file data storehouse server 30;
Audio file shows that server 20 extracts data according to user operation from audio file data storehouse server 30, and by described data display to client;
System management server 40 provides audio file data management service.
Wherein, described audio file data storehouse server 30 comprises:
Audio file library 301, tag library 302, new word storehouse 303, inverse document frequency storehouse 304, user information database 305;
The Word message comprised in the audio file of audio file library 301 saved system management server 40 typing and the content of audio file; Audio file-label relation that label generation server 10 generates; The audio file that user information database generates is played record; Wherein, described audio file-label relation represents the corresponding relation of audio file and corresponding label;
The label of saved system management server 40 typing in tag library 302, keyword, keyword-label relation, and audio file-label relation that label generation server 10 generates; Wherein, described keyword-label relation represents the incidence relation of keyword and corresponding label;
The keyword extracted in the audio file do not comprised in tag library 302 is preserved in new word storehouse 303;
The Word message number comprising keyword is preserved in inverse document frequency storehouse 304;
The user profile of user by client upload and the record of user's displaying audio file is preserved in user information database 305.
Above-mentioned audio file label creation system, by label generation server according to the data genaration audio file label in audio file library and audio file-label relation; Show that server extracts data according to user operation from audio file library by audio file, and by described data display to client; And provide audio file data management service by system management server, provide a kind of intelligentized audio file label creation system, achieve real song label intelligence and generate, efficiency is high, and accuracy is high.
Song label creation system of the present invention and song label generating method one_to_one corresponding of the present invention, the technical characteristic of setting forth in the embodiment of above-mentioned song label generating method and beneficial effect thereof are all applicable to, in the embodiment of song label creation system, hereby state.
Each technical characteristic of the above embodiment can combine arbitrarily, for making description succinct, the all possible combination of each technical characteristic in above-described embodiment is not all described, but, as long as the combination of these technical characteristics does not exist contradiction, be all considered to be the scope that this instructions is recorded.
The above embodiment only have expressed several embodiment of the present invention, and it describes comparatively concrete and detailed, but can not therefore be construed as limiting the scope of the patent.It should be pointed out that for the person of ordinary skill of the art, without departing from the inventive concept of the premise, can also make some distortion and improvement, these all belong to protection scope of the present invention.Therefore, the protection domain of patent of the present invention should be as the criterion with claims.
Claims (10)
1. an audio file label generating method, is characterized in that, comprises the following steps:
By the Word message inputting audio library needing to carry out comprising in the audio file of label setting and the content of audio file, label typing tag library will be commonly used, keyword typing keywords database will be commonly used;
According to the correlation degree of conventional keyword and conventional label, for conventional keyword arranges label, obtain conventional keyword-label relation; Wherein, described keyword-label relation represents the corresponding relation between keyword and corresponding label;
Participle is carried out to described Word message, obtains several words, calculate the TF-IDF value of each word, and according to the acquiring size of TF-IDF value the keyword of audio file; Wherein, described TF-IDF value represents the word frequency of word and the product of inverse document frequency;
The label corresponding according to the keyword of conventional keyword-label relational query audio file; If inquire corresponding label, audio file is associated with corresponding label;
For new word generates new label, and associate described new word and described new label, described new label is set to the label of audio file; Wherein, described new word is the keyword of the audio file not inquiring corresponding label.
2. audio file label generating method according to claim 1, it is characterized in that, described participle is carried out to described Word message, obtain several words, calculate the TF-IDF value of each word, and the step obtaining the keyword of described audio file according to the size of TF-IDF value comprises:
The Word message comprised in the content of audio file is divided into several participles;
Filter the stop words in participle and punctuation mark, obtain useful word;
Calculate word frequency and the inverse document frequency of each useful word;
The TF-IDF value of each word is calculated according to described word frequency and inverse document frequency;
The keyword of audio file according to the acquiring size of described TF-IDF value.
3. audio file label generating method according to claim 2, is characterized in that, the TF-IDF value according to each useful word of following formulae discovery:
tfidf
i,j＝tf
i,j×idf
i，
In formula, tf
i,jrepresent i-th the unduplicated useful word t obtained from a jth audio file
iword frequency; n
i,jrepresent the useful word t obtained from a jth audio file
iquantity; Σ
kn
k,jrepresent the occurrence number sum of all useful word obtained from a jth audio file, idf
irepresent word t
iinverse document frequency, | D| is the sum of the Word message in audio file library, | { j:t
i∈ d
j| for comprising word t in audio file library
ithe quantity of Word message, tfidf
i,jbe i-th unduplicated useful word t
itF-IDF value in a jth audio file.
4. audio file label generating method according to claim 3, is characterized in that, further comprising the steps of:
From inverse document frequency storehouse, inquiry packet is containing go-word language t
iword message;
If do not inquired, then will include and use word t
ithe record of Word message be inserted in inverse document frequency storehouse, and make in inverse document frequency storehouse and comprise word t
ithe quantity of Word message be 1; Otherwise, comprise word t by inverse document frequency storehouse
ithe quantity of Word message add 1.
5. audio file label generating method according to claim 1, is characterized in that, further comprising the steps of:
Whether containing described new word in inquiry new word storehouse;
If contained, then the quantity of new word described in new word storehouse is added 1; Otherwise, described new word is inserted in new word storehouse, and the quantity of this keyword is set to 1.
6. audio file label generating method according to claim 1, is characterized in that, further comprising the steps of:
According to user characteristics, the user of audio file is divided into multiple customer group;
For audio file is set up and associated user group label, the initial value of described customer group label is set to sky; Wherein, described customer group label represents the customer group belonging to user of audio file;
According to the playing duration of audio file in different user group and broadcasting time, the attention rate of statistics audio file in different user group;
Customer group label the highest for attention rate is set to the customer group label of described audio file.
7. audio file label generating method according to claim 6, is characterized in that, described according to the playing duration of audio file in different user group and broadcasting time, and the step of the attention rate of statistics audio file in different user group comprises:
The attention rate of audio file in different user group is added up according to under type:
P
i＝T
i/(T
0,i×T
1)，
In formula, P
irepresent the attention rate of audio file in i-th customer group, T
irepresent the total playing duration of audio file in i-th customer group, T
0, irepresent the broadcasting time of audio file in i-th customer group, T
1represent the duration of audio file.
8. audio file label generating method according to claim 7, is characterized in that, further comprising the steps of:
When user plays described audio file, recording user is to the playing duration of described audio file, described playing duration is accumulated in the playing duration record of audio file described in respective user group, and the broadcasting time of audio file described in respective user group is added 1.
9. an audio file label creation system, is characterized in that, comprising:
Label generation server, audio file show server, audio file data storehouse server, system management server;
Label generation server is audio file correlation tag according to the Word message comprised in the content of the audio file in the server of audio file data storehouse and audio file;
Audio file shows that server extracts the data of audio file from the server of audio file data storehouse according to user operation, and by described data display to client;
System management server provides audio file data management service.
10. audio file label creation system according to claim 9, is characterized in that, described audio file data storehouse server comprises:
Audio file library, tag library, new word storehouse, inverse document frequency storehouse, user information database;
The audio file that audio file-label relation that the Word message comprised in the audio file of audio file library saved system management server typing and the content of audio file, label generation server generate, user information database generate is played record; Wherein, described audio file-label relation represents the corresponding relation of audio file and corresponding label;
The label of saved system management server typing in tag library, keyword, keyword-label relation, and audio file-label relation that label generation server generates; Wherein, described keyword-label relation represents the incidence relation of keyword and corresponding label;
The keyword extracted in the audio file information do not comprised in tag library is preserved in new word storehouse;
The Word message number comprising keyword is preserved in inverse document frequency storehouse;
The record of user profile that user uploads and user's displaying audio file is preserved in user information database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510564305.XA CN105138670B (en) | 2015-09-06 | 2015-09-06 | Audio file label generating method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510564305.XA CN105138670B (en) | 2015-09-06 | 2015-09-06 | Audio file label generating method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105138670A true CN105138670A (en) | 2015-12-09 |
| CN105138670B CN105138670B (en) | 2018-12-14 |
Family
ID=54724017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510564305.XA Active CN105138670B (en) | 2015-09-06 | 2015-09-06 | Audio file label generating method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105138670B (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105718575A (en) * | 2016-01-22 | 2016-06-29 | 华南理工大学 | Crawler-based music labelling method and system |
| CN105740404A (en) * | 2016-01-28 | 2016-07-06 | 上海晶赞科技发展有限公司 | Label association method and device |
| CN105824930A (en) * | 2016-03-17 | 2016-08-03 | 深圳市金立通信设备有限公司 | Voice message processing method and terminal |
| CN105913838A (en) * | 2016-05-19 | 2016-08-31 | 努比亚技术有限公司 | Device and method of audio management |
| CN106294318A (en) * | 2016-08-03 | 2017-01-04 | 浪潮电子信息产业股份有限公司 | A kind of method and device processing educational resource |
| CN107133308A (en) * | 2017-04-28 | 2017-09-05 | 维沃移动通信有限公司 | The single generation method of one kind song and mobile terminal |
| CN107832287A (en) * | 2017-09-26 | 2018-03-23 | 晶赞广告（上海）有限公司 | A kind of label identification method and device, storage medium, terminal |
| CN108959431A (en) * | 2018-06-11 | 2018-12-07 | 中国科学院上海高等研究院 | Label automatic generation method, system, computer readable storage medium and equipment |
| CN109145261A (en) * | 2018-09-04 | 2019-01-04 | 北京奇艺世纪科技有限公司 | A kind of method and apparatus generating label |
| CN109817197A (en) * | 2019-03-04 | 2019-05-28 | 天翼爱音乐文化科技有限公司 | Song generation method, device, computer equipment and storage medium |
| CN111263186A (en) * | 2020-02-18 | 2020-06-09 | 中国传媒大学 | Video generation, playing, searching and processing method, device and storage medium |
| CN111274384A (en) * | 2018-12-05 | 2020-06-12 | 北京京东尚科信息技术有限公司 | Text labeling method and equipment and computer storage medium |
| CN111353071A (en) * | 2018-12-05 | 2020-06-30 | 阿里巴巴集团控股有限公司 | Label generation method and device |
| CN111949776A (en) * | 2020-07-17 | 2020-11-17 | 上海淇馥信息技术有限公司 | Method and device for evaluating user tag and electronic equipment |
| US11558208B2 (en) * | 2019-09-24 | 2023-01-17 | International Business Machines Corporation | Proximity based audio collaboration |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006115718A2 (en) * | 2005-04-25 | 2006-11-02 | Microsoft Corporation | Associating information with an electronic document |
| CN102074235A (en) * | 2010-12-20 | 2011-05-25 | 上海华勤通讯技术有限公司 | Method of video speech recognition and search |
| CN103164471A (en) * | 2011-12-15 | 2013-06-19 | 盛乐信息技术（上海）有限公司 | Recommendation method and system of video text labels |
| CN103207917A (en) * | 2013-04-25 | 2013-07-17 | 百度在线网络技术（北京）有限公司 | Method for marking multimedia content and method and system for generating recommended content |
| CN104090955A (en) * | 2014-07-07 | 2014-10-08 | 科大讯飞股份有限公司 | Automatic audio/video label labeling method and system |
| CN104603773A (en) * | 2012-06-14 | 2015-05-06 | 诺基亚公司 | Method and apparatus for associating interest tags with media items based on social diffusions among users |
-
2015
- 2015-09-06 CN CN201510564305.XA patent/CN105138670B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006115718A2 (en) * | 2005-04-25 | 2006-11-02 | Microsoft Corporation | Associating information with an electronic document |
| CN102074235A (en) * | 2010-12-20 | 2011-05-25 | 上海华勤通讯技术有限公司 | Method of video speech recognition and search |
| CN103164471A (en) * | 2011-12-15 | 2013-06-19 | 盛乐信息技术（上海）有限公司 | Recommendation method and system of video text labels |
| CN104603773A (en) * | 2012-06-14 | 2015-05-06 | 诺基亚公司 | Method and apparatus for associating interest tags with media items based on social diffusions among users |
| CN103207917A (en) * | 2013-04-25 | 2013-07-17 | 百度在线网络技术（北京）有限公司 | Method for marking multimedia content and method and system for generating recommended content |
| CN104090955A (en) * | 2014-07-07 | 2014-10-08 | 科大讯飞股份有限公司 | Automatic audio/video label labeling method and system |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105718575B (en) * | 2016-01-22 | 2019-01-29 | 华南理工大学 | Patch music label method and system based on crawler |
| CN105718575A (en) * | 2016-01-22 | 2016-06-29 | 华南理工大学 | Crawler-based music labelling method and system |
| CN105740404A (en) * | 2016-01-28 | 2016-07-06 | 上海晶赞科技发展有限公司 | Label association method and device |
| CN105824930A (en) * | 2016-03-17 | 2016-08-03 | 深圳市金立通信设备有限公司 | Voice message processing method and terminal |
| CN105913838A (en) * | 2016-05-19 | 2016-08-31 | 努比亚技术有限公司 | Device and method of audio management |
| CN106294318A (en) * | 2016-08-03 | 2017-01-04 | 浪潮电子信息产业股份有限公司 | A kind of method and device processing educational resource |
| CN107133308B (en) * | 2017-04-28 | 2020-01-14 | 维沃移动通信有限公司 | Singing list generation method and mobile terminal |
| CN107133308A (en) * | 2017-04-28 | 2017-09-05 | 维沃移动通信有限公司 | The single generation method of one kind song and mobile terminal |
| CN107832287A (en) * | 2017-09-26 | 2018-03-23 | 晶赞广告（上海）有限公司 | A kind of label identification method and device, storage medium, terminal |
| CN108959431A (en) * | 2018-06-11 | 2018-12-07 | 中国科学院上海高等研究院 | Label automatic generation method, system, computer readable storage medium and equipment |
| CN108959431B (en) * | 2018-06-11 | 2022-07-05 | 中国科学院上海高等研究院 | Automatic label generation method, system, computer readable storage medium and equipment |
| CN109145261A (en) * | 2018-09-04 | 2019-01-04 | 北京奇艺世纪科技有限公司 | A kind of method and apparatus generating label |
| CN109145261B (en) * | 2018-09-04 | 2022-12-06 | 北京奇艺世纪科技有限公司 | Method and device for generating label |
| CN111274384B (en) * | 2018-12-05 | 2024-02-06 | 北京京东尚科信息技术有限公司 | Text labeling method, equipment and computer storage medium thereof |
| CN111274384A (en) * | 2018-12-05 | 2020-06-12 | 北京京东尚科信息技术有限公司 | Text labeling method and equipment and computer storage medium |
| CN111353071A (en) * | 2018-12-05 | 2020-06-30 | 阿里巴巴集团控股有限公司 | Label generation method and device |
| CN109817197A (en) * | 2019-03-04 | 2019-05-28 | 天翼爱音乐文化科技有限公司 | Song generation method, device, computer equipment and storage medium |
| US11558208B2 (en) * | 2019-09-24 | 2023-01-17 | International Business Machines Corporation | Proximity based audio collaboration |
| CN111263186A (en) * | 2020-02-18 | 2020-06-09 | 中国传媒大学 | Video generation, playing, searching and processing method, device and storage medium |
| CN111949776A (en) * | 2020-07-17 | 2020-11-17 | 上海淇馥信息技术有限公司 | Method and device for evaluating user tag and electronic equipment |
| CN111949776B (en) * | 2020-07-17 | 2023-09-22 | 上海淇馥信息技术有限公司 | User tag evaluation method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105138670B (en) | 2018-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105138670A (en) | Audio file label generation method and system | |
| CN102982153B (en) | A kind of information retrieval method and device thereof | |
| US9990368B2 (en) | System and method for automatic generation of information-rich content from multiple microblogs, each microblog containing only sparse information | |
| CN103390051B (en) | A kind of topic detection and tracking method based on microblog data | |
| US9317550B2 (en) | Query expansion | |
| CN103853834B (en) | Text structure analysis-based Web document abstract generation method | |
| CN107577688A (en) | Original article influence power analysis system based on media information collection | |
| US20140172415A1 (en) | Apparatus, system, and method of providing sentiment analysis result based on text | |
| US20150205580A1 (en) | Method and System for Sorting Online Videos of a Search | |
| CN102033919A (en) | Method and system for extracting text key words | |
| CN102682120B (en) | Method and device for acquiring essential article commented on network | |
| CN103778200B (en) | A kind of message information source abstracting method and its system | |
| CN102043843A (en) | Method and obtaining device for obtaining target entry based on target application | |
| CN103853722A (en) | Query based keyword extension method, device and system | |
| CN105302810A (en) | Information search method and apparatus | |
| CN102890689A (en) | Method and system for building user interest model | |
| CN104077417A (en) | Figure tag recommendation method and system in social network | |
| CN103365910A (en) | Method and system for information retrieval | |
| CN105573995A (en) | Interest identification method, interest identification equipment and data analysis method | |
| CN103177036A (en) | Method and system for label automatic extraction | |
| CN111026965B (en) | Hot topic tracing method and device based on knowledge graph | |
| CN106126605B (en) | Short text classification method based on user portrait | |
| CN103488782B (en) | A kind of method utilizing lyrics identification music emotion | |
| CN104615627A (en) | Event public sentiment information extracting method and system based on micro-blog platform | |
| CN103778122A (en) | Searching method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB03 | Change of inventor or designer information |
Inventor after: Lu Zanxin Inventor after: Zhu Yingbo Inventor after: Zeng Rong Inventor after: Wang Wei Inventor before: Lu Zanxin |
|
| COR | Change of bibliographic data | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |