CN103927366A - Method and system for automatically playing songs according to pictures - Google Patents
Method and system for automatically playing songs according to pictures Download PDFInfo
- Publication number
- CN103927366A CN103927366A CN201410161263.0A CN201410161263A CN103927366A CN 103927366 A CN103927366 A CN 103927366A CN 201410161263 A CN201410161263 A CN 201410161263A CN 103927366 A CN103927366 A CN 103927366A
- Authority
- CN
- China
- Prior art keywords
- picture
- label
- lyrics
- theme
- module
- 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
- 238000000034 method Methods 0.000 title claims abstract description 27
- 239000012634 fragment Substances 0.000 claims description 9
- 238000002372 labelling Methods 0.000 claims description 9
- 238000003064 k means clustering Methods 0.000 claims description 4
- 238000009826 distribution Methods 0.000 description 5
- 239000000203 mixture Substances 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003058 natural language processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000004615 ingredient Substances 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000002156 mixing Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/683—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/685—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using automatically derived transcript of audio data, e.g. lyrics
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Library & Information Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides a method and system for automatically playing songs according to pictures. The method comprises the following steps that S1, all the pictures in a picture library are marked with labels, and lyrics of all the songs in the song library are marked with labels; S2, all the pictures in the picture library are preprocessed to obtain a dictionary, and all the pictures in the picture library are reconstructed according to the dictionary; S3, the reconstructed pictures are scanned, subjects of the pictures are obtained through LDA, the labels corresponding to the pictures are added to the subjects of the pictures, subjects of the lyrics in the song library are obtained through the LDA, and the labels corresponding to the lyrics are added to the subjects of the lyrics; S4, the labels of the pictures are obtained according to the subjects of the pictures in the picture library, the lyrics with the labels which are the same as the labels of the pictures are found in the song library, and the corresponding songs are played.
Description
Technical field
The invention belongs to machine learning and area of pattern recognition, be specifically related to a kind of according to the method and system of the automatic played songs of picture.
Background technology
Along with informationalized universal rapidly, the sharply increase that the information of all kinds (as text, audio frequency, video etc.) has been brought quantity of information.How by content is close but the information that type is different integrate, it is classified according to content, make people when the information of a certain certain content of demand, needn't be limited to the type of information, thereby reach the object of diversification acquisition of information, become at present problem demanding prompt solution.
In order to process these dissimilar information, need to carry out the classification based on topic model to various types of information.Wherein, LDA (Latent Dirichlet Allocation) topic model algorithm is the method that the implicit theme of word is carried out to modeling, it is a set probability model, mainly for the treatment of discrete data acquisition, mainly be used at present in data mining and natural language processing, can reduce dimension.It has overcome the shortcoming of document similarity calculating method in conventional information retrieval, and can in magnanimity internet data, find out the semantic topic between word by automatic seeking.
In addition, the potential semantic indexing of PLSI(probability, probabilistic latent semantic indexing) be a kind of conventional probability semantic model, it is modeled as each word in document in the sample taking out in mixture model.Wherein, the composition in mixture model is the stochastic variable that meets multinomial distribution.In document, different words may be to be produced by different semantemes.Each document is represented as a series of mixing ingredient, so dimensionality reduction has become in a fixed set probability distribution semantically.The generative process of LDA model is identical with PLSI in itself.PLSI obtains a mixed decomposition from potential class model, the retrieval that it is widely used in information at present and filtration, natural language processing, the field that machine learning is relevant with some.The main difference of LDA and PLSI is: LDA document the distribution on theme and the distribution of theme on word list all regard stochastic variable as, by super ginseng, controlled; And PLSI regards these two distributions as parameter as.This causes the increase along with training number of files, and in test set, the model being drawn by training set may cause the phenomenon of over-fitting.
Current existing a lot of scholars are on the basis of potential Di Li Cray model (being LDA), by adding relevance factors between label, time factor, author, supervision or without supervision factor etc., having made further improvement.At present, the application of topic model has covered a plurality of fields such as text mining, information processing, image processing.
In view of the foregoing, the present invention utilizes LDA topic model to have analog information content but the information of different information types (picture and text) is classified, and provides a kind of according to the method and system of the automatic played songs of picture.Make user to listen attentively to the music close with picture situation in appreciation picture limit, limit, improved people's living standard and sentiment mastery.And children can see that picture limit is heard and the song of the corresponding lyrics of picture theme in limit, the understanding of increase to things, the present invention carries out child's early education to father and mother and has positive meaning.
Summary of the invention
The invention provides a kind ofly according to the method for the automatic played songs of picture, comprise the following steps:
S1, to all pictures mark labels in picture library, the lyrics mark label to all songs in song storehouse;
S2, all pictures in picture library are carried out to pre-service obtain dictionary, according to all pictures in picture library described in described dictionary reconstruct;
S3, scan the picture of reconstruct, by LDA, obtain the theme of described picture, label corresponding to described picture added in the theme of himself, by described LDA, obtain the theme of the lyrics in song storehouse simultaneously, and label corresponding to the described lyrics added in the theme of himself;
S4, according to the theme of picture in picture library, obtain the label of described picture, in song storehouse, find the lyrics that label is identical with the label of described picture, and play corresponding song.
Preferably, the label in step S1, described picture and the lyrics being marked is respectively to belong to same set.
Preferably, described in step S2, pretreated process is: in the every pictures from picture library, select N fragment, the SIFT vector representation of each fragment, from all fragments, with k-means clustering algorithm, obtain T central point, described T central point forms dictionary.
Preferably, in step S3, if picture or the lyrics do not have corresponding label, the label using the label of described picture or lyrics theme as described picture or the lyrics.
Preferably, the LDA that described in step S3, picture and the lyrics pass through is identical topic model algorithm.
Preferably, a label only belongs to a theme.
It is a kind of according to the system of the automatic played songs of picture that the present invention also provides, comprise label for labelling module, picture pretreatment module, cluster module and matching module, described label for labelling module connects picture pretreatment module, described picture pretreatment module connects cluster module, described cluster module matching connection module, described matching module connects player, described cluster module, comprise scanner, LDA picture cluster module and LDA text cluster module, described scanner connects LDA picture cluster module.Described label for labelling module, for to all picture marks of picture library label, marks label to the lyrics of all songs in song storehouse.Described picture pretreatment module, obtains dictionary for all pictures of picture library are carried out to pre-service, and according to all pictures in picture library described in described dictionary reconstruct.Described cluster module, for scanning after the picture of reconstruct, by LDA, obtain the theme of described picture, label corresponding to described picture added in the theme of himself, by described LDA, obtain the theme of the lyrics in song storehouse simultaneously, and label corresponding to the described lyrics added in the theme of himself.Described matching module for obtain the label of described picture according to the theme of picture library picture, finds the lyrics that label is identical with the label of described picture in song storehouse.Described player, for playing the lyrics corresponding song identical with the label of described picture.
By provided by the invention according to the method and system of the automatic played songs of picture, first the lyrics of song in the picture in picture library and song storehouse are marked to label, then picture is carried out to pre-service and obtain dictionary reconstructed picture, thereby carry out place mat for LDA obtains picture theme automatically.Obtain after picture theme, label corresponding to picture added in the theme of himself, by LDA, obtain the theme of the lyrics in song storehouse simultaneously, and label corresponding to the lyrics added in the theme of himself, finally according to the theme of picture in picture library, obtain its label, in song storehouse, find the lyrics that label is identical and play corresponding song.So, when finding new picture, can automatically play the music close with picture situation by native system, for children are early provided by a kind of mode easily that provides.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.
Fig. 1 be preferred embodiment of the present invention provide according to the method flow diagram of the automatic played songs of picture;
Fig. 2 be preferred embodiment of the present invention provide according to the system schematic of the automatic played songs of picture.
Embodiment
Hereinafter with reference to accompanying drawing, also describe the present invention in detail in conjunction with the embodiments.It should be noted that, in the situation that not conflicting, embodiment and the feature in embodiment in the application can combine mutually.
Fig. 1 be preferred embodiment of the present invention provide according to the method flow diagram of the automatic played songs of picture.As shown in Figure 1, the method according to the automatic played songs of picture that preferred embodiment of the present invention provides comprises step S1 to S4.
Step S1: to all picture mark labels in picture library, the lyrics mark label to all songs in song storehouse.
Particularly, the samples pictures in picture library is that part is through mark, described label such as tree, the sun or meadow etc., the song in existing song storehouse is processed, to the processing of song, mainly according to the lyrics, and the lyrics are exactly text meanwhile.Described song also part is marked.
In this, the label that described picture and the lyrics are marked is respectively to belong to same set.Such as tree, the sun or meadow etc.
Step S2: all pictures in picture library are carried out to pre-service and obtain dictionary, according to all pictures in picture library described in described dictionary reconstruct.
Particularly, in the present embodiment, SIFT for picture (Scale-invariant feature transform, the conversion of yardstick invariant features) extracted vector feature.Pretreated process is: in the every pictures from picture library, select N fragment, the SIFT vector representation of each fragment, from all fragments, obtains T central point with k-means clustering algorithm, and described T central point forms dictionary.Conventionally, k-means clustering algorithm adopts distance as the evaluation index of similarity, thinks that the distance of two objects is nearer, and its similarity is just larger.
Step S3: the picture that scans reconstruct, by LDA, obtain the theme of described picture, label corresponding to described picture added in the theme of himself, by described LDA, obtain the theme of the lyrics in song storehouse simultaneously, and label corresponding to the described lyrics added in the theme of himself.
Particularly, after all pictures in dictionary restructuring graph valut, the input using all pictures as LDA.If set theme number K, the output of LDA training pattern is picture theme and distributes, and theme dictionary distributes.
For example, when scanning a new picture, be first reconstructed into the expression of dictionary, then put in LDA and test, obtain picture theme and distribute, the theme that in choosing a topic distributes, probability is large is as the explanation of this picture.The label of this width picture is added in the middle of this theme, now just obtain the label of the represented theme of this picture, i.e. the represented content of this picture.
In the present embodiment, if picture or the lyrics do not have corresponding label, the label using the label of described picture or lyrics theme as described picture or the lyrics.In this, a label only belongs to a theme.
In addition the LDA that, described picture and the lyrics pass through is identical topic model algorithm.
Step S4: obtain the label of described picture according to the theme of picture in picture library, find the lyrics that label is identical with the label of described picture in song storehouse, and play corresponding song.
Particularly, when picture has marked after label by LDA, it is mated with corresponding lyrics label in song storehouse, in the corresponding song of the label matching, randomly draw a song and play by player.
Fig. 2 be preferred embodiment of the present invention provide according to the system schematic of the automatic played songs of picture.As shown in Figure 2, the system according to the automatic played songs of picture that preferred embodiment of the present invention provides comprises label for labelling module 1, picture pretreatment module 2, cluster module 3 and matching module 4, described label for labelling module 1 connects picture pretreatment module 2, described picture pretreatment module 2 connects cluster module 3, described cluster module 3 matching connection modules 4, described matching module 4 connects player 5.Described cluster module 3, comprises scanner 31, LDA picture cluster module 32 and LDA text cluster module 33, and described scanner 31 connects LDA picture cluster module 32.Described label for labelling module 1, for to all picture marks of picture library label, marks label to the lyrics of all songs in song storehouse.Described picture pretreatment module 2, obtains dictionary for all pictures of picture library are carried out to pre-service, and according to all pictures in picture library described in described dictionary reconstruct.Described cluster module 3, for scanning after the picture of reconstruct, by LDA, obtain the theme of described picture, label corresponding to described picture added in the theme of himself, by described LDA, obtain the theme of the lyrics in song storehouse simultaneously, and label corresponding to the described lyrics added in the theme of himself.Described matching module 4 for obtain the label of described picture according to the theme of picture library picture, finds the lyrics that label is identical with the label of described picture in song storehouse.Described player 5, for playing the lyrics corresponding song identical with the label of described picture.Operating process about said system is similar with said method, therefore repeat no more in this.
In sum, according to the present invention, preferred embodiment provides according to the method and system of the automatic played songs of picture, first the lyrics of song in the picture in picture library and song storehouse are marked to label, then picture is carried out to pre-service and obtain dictionary reconstructed picture, thereby carry out place mat for LDA obtains picture theme automatically.Obtain after picture theme, label corresponding to picture added in the theme of himself, by LDA, obtain the theme of the lyrics in song storehouse simultaneously, and label corresponding to the lyrics added in the theme of himself, finally according to the theme of picture in picture library, obtain its label, in song storehouse, find the lyrics that label is identical and play corresponding song.So, can automatically play the song of corresponding bent wind according to image content, improve people's living standard and sentiment mastery.
Above-mentioned explanation to the disclosed embodiments, makes professional and technical personnel in the field can realize or use the present invention.To the multiple modification of these embodiment, will be apparent for those skilled in the art, General Principle as defined herein can, in the situation that not departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention will can not be restricted to embodiment illustrated herein, but will meet the widest scope consistent with principle disclosed herein and features of novelty.Above-mentioned explanation to the disclosed embodiments, makes professional and technical personnel in the field can realize or use the present invention.To the multiple modification of these embodiment, will be apparent for those skilled in the art, General Principle as defined herein can, in the situation that not departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention will can not be restricted to embodiment illustrated herein, but will meet the widest scope consistent with principle disclosed herein and features of novelty.
Claims (7)
1. according to a method for the automatic played songs of picture, it is characterized in that, comprise the following steps:
S1, to all pictures mark labels in picture library, the lyrics mark label to all songs in song storehouse;
S2, all pictures in picture library are carried out to pre-service obtain dictionary, according to all pictures in picture library described in described dictionary reconstruct;
S3, scan the picture of reconstruct, by LDA, obtain the theme of described picture, label corresponding to described picture added in the theme of himself, by described LDA, obtain the theme of the lyrics in song storehouse simultaneously, and label corresponding to the described lyrics added in the theme of himself;
S4, according to the theme of picture in picture library, obtain the label of described picture, in song storehouse, find the lyrics that label is identical with the label of described picture, and play corresponding song.
2. method according to claim 1, is characterized in that, the label in step S1, described picture and the lyrics being marked is respectively to belong to same set.
3. method according to claim 1, it is characterized in that, described in step S2, pretreated process is: in the every pictures from picture library, select N fragment, the SIFT vector representation of each fragment, from all fragments, with k-means clustering algorithm, obtain T central point, described T central point forms dictionary.
4. method according to claim 1, is characterized in that, in step S3, if picture or the lyrics do not have corresponding label, and the label using the label of described picture or lyrics theme as described picture or the lyrics.
5. method according to claim 1, is characterized in that, the LDA that picture described in step S3 and the lyrics pass through is identical topic model algorithm.
6. method according to claim 1, is characterized in that, a label only belongs to a theme.
7. one kind according to the system of the automatic played songs of picture, it is characterized in that, comprise label for labelling module, picture pretreatment module, cluster module, matching module and player, described label for labelling module connects picture pretreatment module, described picture pretreatment module connects cluster module, described cluster module matching connection module, described matching module connects player, described cluster module, comprise scanner, LDA picture cluster module and LDA text cluster module, described scanner connects LDA picture cluster module
Described label for labelling module, for to all picture marks of picture library label, marks label to the lyrics of all songs in song storehouse;
Described picture pretreatment module, obtains dictionary for all pictures of picture library are carried out to pre-service, and according to all pictures in picture library described in described dictionary reconstruct;
Described cluster module, for scanning after the picture of reconstruct, by LDA, obtain the theme of described picture, label corresponding to described picture added in the theme of himself, by described LDA, obtain the theme of the lyrics in song storehouse simultaneously, and label corresponding to the described lyrics added in the theme of himself;
Described matching module for obtain the label of described picture according to the theme of picture library picture, finds the lyrics that label is identical with the label of described picture in song storehouse;
Described player, for playing the lyrics corresponding song identical with the label of described picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410161263.0A CN103927366B (en) | 2014-04-21 | 2014-04-21 | Method and system for automatically playing songs according to pictures |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410161263.0A CN103927366B (en) | 2014-04-21 | 2014-04-21 | Method and system for automatically playing songs according to pictures |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103927366A true CN103927366A (en) | 2014-07-16 |
| CN103927366B CN103927366B (en) | 2017-03-22 |
Family
ID=51145587
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410161263.0A Expired - Fee Related CN103927366B (en) | 2014-04-21 | 2014-04-21 | Method and system for automatically playing songs according to pictures |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103927366B (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104599307A (en) * | 2015-02-11 | 2015-05-06 | 成都品果科技有限公司 | Mobile terminal animated image display method |
| CN105930485A (en) * | 2016-04-28 | 2016-09-07 | 深圳市金立通信设备有限公司 | Audio media playing method, communication device and network system |
| CN107918614A (en) * | 2016-10-08 | 2018-04-17 | 北京小唱科技有限公司 | A kind of recommendation method and server for singing accompaniment |
| CN108920648A (en) * | 2018-07-03 | 2018-11-30 | 四川大学 | It is a kind of based on music-image, semantic relationship across mode matching method |
| WO2019062716A1 (en) * | 2017-09-30 | 2019-04-04 | 腾讯科技（深圳）有限公司 | Method and apparatus for generating music |
| CN109903783A (en) * | 2019-02-27 | 2019-06-18 | 百度在线网络技术（北京）有限公司 | Multimedia control method, device and terminal |
| CN112182281A (en) * | 2019-07-05 | 2021-01-05 | 腾讯科技（深圳）有限公司 | Audio recommendation method and device and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008056872A1 (en) * | 2006-11-09 | 2008-05-15 | Kyoung Jun Lee | Contents generation equipment and method compounding rf tag information and contents |
| CN101312565A (en) * | 2007-05-25 | 2008-11-26 | 上海美通无线网络信息有限公司 | Method for synchronous lyrics browsing when playing music on mobile phone |
| CN101567203A (en) * | 2008-04-24 | 2009-10-28 | 深圳富泰宏精密工业有限公司 | System and method for automatically searching and playing music |
| US20120124054A1 (en) * | 2004-11-01 | 2012-05-17 | Sony Corporation | Recording medium, recording device, recording method, data search device, data search method, and data generating device |
| US20120166414A1 (en) * | 2008-08-11 | 2012-06-28 | Ultra Unilimited Corporation (dba Publish) | Systems and methods for relevance scoring |
| CN103677776A (en) * | 2012-09-19 | 2014-03-26 | 腾讯科技（深圳）有限公司 | Music player and automatic skin switching method and system thereof |
-
2014
- 2014-04-21 CN CN201410161263.0A patent/CN103927366B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120124054A1 (en) * | 2004-11-01 | 2012-05-17 | Sony Corporation | Recording medium, recording device, recording method, data search device, data search method, and data generating device |
| WO2008056872A1 (en) * | 2006-11-09 | 2008-05-15 | Kyoung Jun Lee | Contents generation equipment and method compounding rf tag information and contents |
| CN101312565A (en) * | 2007-05-25 | 2008-11-26 | 上海美通无线网络信息有限公司 | Method for synchronous lyrics browsing when playing music on mobile phone |
| CN101567203A (en) * | 2008-04-24 | 2009-10-28 | 深圳富泰宏精密工业有限公司 | System and method for automatically searching and playing music |
| US20120166414A1 (en) * | 2008-08-11 | 2012-06-28 | Ultra Unilimited Corporation (dba Publish) | Systems and methods for relevance scoring |
| CN103677776A (en) * | 2012-09-19 | 2014-03-26 | 腾讯科技（深圳）有限公司 | Music player and automatic skin switching method and system thereof |
Non-Patent Citations (1)
| Title |
|---|
| 刘凯鹏，方滨兴: "一种基于社会性标注的网页排序算法", 《计算机学报》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104599307A (en) * | 2015-02-11 | 2015-05-06 | 成都品果科技有限公司 | Mobile terminal animated image display method |
| CN104599307B (en) * | 2015-02-11 | 2018-05-15 | 成都品果科技有限公司 | A kind of method for showing picture using animation on mobile terminals |
| CN105930485A (en) * | 2016-04-28 | 2016-09-07 | 深圳市金立通信设备有限公司 | Audio media playing method, communication device and network system |
| CN107918614A (en) * | 2016-10-08 | 2018-04-17 | 北京小唱科技有限公司 | A kind of recommendation method and server for singing accompaniment |
| WO2019062716A1 (en) * | 2017-09-30 | 2019-04-04 | 腾讯科技（深圳）有限公司 | Method and apparatus for generating music |
| US11301641B2 (en) | 2017-09-30 | 2022-04-12 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for generating music |
| CN108920648A (en) * | 2018-07-03 | 2018-11-30 | 四川大学 | It is a kind of based on music-image, semantic relationship across mode matching method |
| CN108920648B (en) * | 2018-07-03 | 2021-06-22 | 四川大学 | Cross-modal matching method based on music-image semantic relation |
| CN109903783A (en) * | 2019-02-27 | 2019-06-18 | 百度在线网络技术（北京）有限公司 | Multimedia control method, device and terminal |
| CN112182281A (en) * | 2019-07-05 | 2021-01-05 | 腾讯科技（深圳）有限公司 | Audio recommendation method and device and storage medium |
| CN112182281B (en) * | 2019-07-05 | 2023-09-19 | 腾讯科技（深圳）有限公司 | Audio recommendation method, device and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103927366B (en) | 2017-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Hande et al. | KanCMD: Kannada CodeMixed dataset for sentiment analysis and offensive language detection | |
| CN103927366A (en) | Method and system for automatically playing songs according to pictures | |
| Bruni et al. | Multimodal distributional semantics | |
| US10248865B2 (en) | Identifying presentation styles of educational videos | |
| Habibian et al. | Recommendations for video event recognition using concept vocabularies | |
| Cheung et al. | Sequence clustering and labeling for unsupervised query intent discovery | |
| Sperberg‐McQueen | Classification and its Structures | |
| Gupta et al. | Vico: Word embeddings from visual co-occurrences | |
| CN111046194A (en) | Method for constructing multi-mode teaching knowledge graph | |
| CN103559193A (en) | Topic modeling method based on selected cell | |
| CN112559781B (en) | Image retrieval system and method | |
| Li et al. | Share-and-chat: Achieving human-level video commenting by search and multi-view embedding | |
| Jacobs | Introduction: Text power and intelligent systems | |
| Liu et al. | Discoursing disciplinarity: A bibliometric analysis of published research in the past 30 years | |
| Chatbri et al. | Automatic MOOC video classification using transcript features and convolutional neural networks | |
| Prasetyo et al. | Implementation of K-Means Clustering for Analysis Students English Proficiency | |
| JP5972096B2 (en) | Apparatus, method and program for extracting posts related to contents | |
| Kammoun et al. | ServOMap results for OAEI 2013. | |
| Kotsakis et al. | Extending radio broadcasting semantics through adaptive audio segmentation automations | |
| de Boer et al. | Improving video event retrieval by user feedback | |
| Fourati et al. | Automatic audiovisual documents genre description | |
| Saito et al. | A Framework for Developing Presentation Skill with Presentation Schema | |
| Wang et al. | Music Recommendation Based on “User-Points-Music” Cascade Model and Time Attenuation Analysis | |
| Timmaraju et al. | Addressing Challenges in Building Web-Scale Content Classification Systems | |
| McParlane et al. | Collections for automatic image annotation and photo tag recommendation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170322 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |