CN105893463B - Album input method and device - Google Patents
Album input method and device Download PDFInfo
- Publication number
- CN105893463B CN105893463B CN201610173450.XA CN201610173450A CN105893463B CN 105893463 B CN105893463 B CN 105893463B CN 201610173450 A CN201610173450 A CN 201610173450A CN 105893463 B CN105893463 B CN 105893463B
- Authority
- CN
- China
- Prior art keywords
- audio
- video
- album
- similar
- library
- 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.)
- Active
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
- G06F16/60—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
- G06F16/686—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings
-
- 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
- G06F16/61—Indexing; Data structures therefor; Storage structures
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of album input method and devices, belong to data input technical field.The described method includes: by obtaining album information；For each audio-video in n audio-video, detects and whether there is similar audio-video similar to audio-video in audio-video library；There are the numbers of the audio-video of similar audio-video in n audio-video of statistics；Whether detection number reaches first threshold；If not up to first threshold, by album information typing to audio-video library；When solving the album that same names are not present in backstage manager in finding audio repository, the song in the album name and album is subjected to typing, leads to the different albums that may there are problems that same song in audio repository；Reached only when n audio-video in audio-video library there are the number of the audio-video of similar audio-video be less than first threshold when, just album information is entered into audio-video library, avoids the different albums that there are problems that same song in audio-video library.
Description
Technical field
The present invention relates to data input technical field, in particular to a kind of album input method and device.
Background technique
Music player is Internet application very popular with users at present.
In the prior art, it is flat to artificially collect major music in order to provide the user with more audio resources by backstage manager
The audio resource of platform, and manually by the audio resource typing of collection into audio repository.Extremely with backstage manager's typing album
For audio repository, backstage manager can manually by album name, album song and corresponding singer be entered into sound
In frequency library.Specifically, whether backstage manager first inquires the singer of the album in audio repository in audio repository in Input Process
Singer's list in, when being present in singer's list, then inquire in audio repository with the presence or absence of identical album name, when being not present
When identical album name, then the song in the album name and the album is entered into audio repository.
In the implementation of the present invention, the inventor finds that the existing technology has at least the following problems:
When the album of same names is not present in backstage manager in finding audio repository, by the album name and album
In song carry out typing, this results in the different albums that may have same song in audio repository.
Summary of the invention
When the album of same names is not present in order to solve backstage manager in finding audio repository, by the album name
And the song in album carries out typing, leads to the different albums that may there are problems that same song in audio repository, this hair
Bright embodiment provides a kind of album input method and device.The technical solution is as follows:
In a first aspect, a kind of album input method is provided, this method comprises:
Album information is obtained, album information includes n audio-video in target album, and n is positive integer；
For each audio-video in n audio-video, detect in audio-video library with the presence or absence of similar similar to audio-video
Audio-video；
There are the numbers of the audio-video of similar audio-video in n audio-video of statistics；
Whether detection number reaches first threshold, and first threshold is greater than 0 and is less than or equal to n；
If not up to first threshold, by album information typing to audio-video library.
In one possible implementation, for each audio-video in n audio-video, detect in audio-video library whether
In the presence of similar audio-video similar to audio-video, comprising:
The audio-video classification in audio-video library is obtained, each audio-video in each audio-video classification is audio/video fingerprint
Similarity is higher than the similar audio-video of second threshold；
For each audio-video in n audio-video, the audio/video fingerprint of audio-video is obtained；
For each audio-video in n audio-video, by the part sound in audio/video fingerprint and the classification of each audio-video
The audio/video fingerprint of video is compared, and detects and whether there is similar audio-video similar to audio-video in audio-video library.
In one possible implementation, for each audio-video in n audio-video, by audio/video fingerprint and respectively
The audio/video fingerprint of part audio-video in the classification of a audio-video is compared, and detect whether there is and audio-video in audio-video library
Similar similar audio-video, comprising:
For i-th of audio-video classification in the classification of each audio-video, calculate in audio/video fingerprint and part audio-video
Similarity between the audio/video fingerprint of each audio-video, 1≤i≤N, N are the total number of the audio-video classification in audio-video library；
If there is the similarity more than second threshold in each similarity being calculated, it is determined that audio-video exists in library
Similar audio-video similar to audio-video；
If there is no the similarities more than second threshold to enable i=i+ in i < N in each similarity being calculated
1, it executes calculate the similarity between audio/video fingerprint and the audio/video fingerprint of each audio-video in the audio-video of part again
Step.
In one possible implementation, this method, further includes:
If reaching first threshold, whether the number for detecting the similar audio-video in same album reaches first threshold；
If reaching first threshold, it is determined that target album belongs to the album that detection obtains.
In one possible implementation, album information further includes the album name of target album；This method further include:
It detects and whether there is the album of the same name with album name in audio-video library；
If testing result is that there is no whether the number for executing the similar audio-video in the same album of detection reaches first
The step of threshold value.
In one possible implementation, target album is album of songs, and n audio-video is n song；This method is also
Include:
Obtain target singer corresponding to target album；
For every song in n song, the affiliated song of song in audio-video library is obtained according to the song fingerprints of song
Singer corresponding to every song in classification；
Whether detection target singer is a member in each singer got；
If testing result is yes, it is determined that target singer has existed and audio-video library.
In one possible implementation, this method further include:
If testing result be it is no, obtained according to the song fingerprints of song each in other categorizing songs in audio-video library
Singer corresponding to song；
Whether detection target singer is a member in singer corresponding to each song in other categorizing songs；
If testing result be it is no, by target singer typing to audio-video.
Second aspect, provides a kind of album input device, which includes:
Album obtains module, and for obtaining album information, album information includes n audio-video in target album, and n is positive
Integer；
Approx imately-detecting module, for for each audio-video in n audio-video, detect in audio-video library with the presence or absence of with
The similar similar audio-video of audio-video；
Number statistical module, for counting, there are the numbers of the audio-video of similar audio-video in n audio-video；
Number detection module, for detecting whether number reaches first threshold, first threshold is greater than 0 and is less than or equal to n；
Album recording module is used in not up to first threshold, by album information typing to audio-video library.
In one possible implementation, approx imately-detecting module, comprising:
Classification acquisition submodule, it is each in each audio-video classification for obtaining the classification of the audio-video in audio-video library
Audio-video is that the similarity of audio/video fingerprint is higher than the similar audio-video of second threshold；
Fingerprint acquisition submodule, for for each audio-video in n audio-video, the audio-video for obtaining audio-video to refer to
Line；
Fingerprint comparison submodule, for for each audio-video in n audio-video, by audio/video fingerprint and each sound
The audio/video fingerprint of part audio-video in visual classification is compared, and detects in audio-video library with the presence or absence of similar to audio-video
Similar audio-video.
In one possible implementation, fingerprint comparison submodule, comprising:
Similar computing unit calculates audio/video fingerprint for classifying for i-th of audio-video in the classification of each audio-video
Similarity between the audio/video fingerprint of each audio-video in the audio-video of part, 1≤i≤N, N are the sound in audio-video library
The total number of visual classification；
Similar determination unit, when for there is the similarity more than second threshold in each similarity being calculated,
Determine there is similar audio-video similar to audio-video in audio-video library；
Similar computing unit is also used to that the similarity more than second threshold is not present in each similarity being calculated
When, in i < N, i=i+1 is enabled, executes the audio-video for calculating each audio-video in audio/video fingerprint and part audio-video again
The step of similarity between fingerprint.
In one possible implementation, the device, further includes:
Album detection module, the number for when reaching first threshold, detecting the similar audio-video in same album are
It is no to reach first threshold；
Album determining module, for when reaching first threshold, determining that target album belongs to the album that detection obtains.
In one possible implementation, album information further includes the album name of target album；
The device, further includes:
Title detection module, for detecting in audio-video library with the presence or absence of the album of the same name with album name；
Album detection module, for executing the similar audio-video detected in same album in the absence of testing result is
Number the step of whether reaching first threshold.
In one possible implementation, target album is album of songs, and n audio-video is n song；
The device, further includes:
First obtains module, for obtaining target singer corresponding to target album；
Second obtains module, for obtaining audio-video according to the song fingerprints of song for every song in n song
Singer corresponding to every song in library in the affiliated categorizing songs of song；
First detection module, for detecting whether target singer is a member in each singer got
Singer's determining module, for determining that target singer has existed and audio-video library when testing result, which is, is.
In one possible implementation, the device, further includes:
Third obtains module, for obtaining its in audio-video library according to the song fingerprints of song when testing result is no
Singer corresponding to each song in its categorizing songs；
Second detection module, for detecting whether target singer is song corresponding to each song in other categorizing songs
A member in hand
Singer's recording module is used for when testing result is no, by target singer typing to audio-video.
Technical solution provided in an embodiment of the present invention has the benefit that
By obtaining album information, album information includes n audio-video in target album, and n is positive integer；For n
Each audio-video in audio-video detects and whether there is similar audio-video similar to audio-video in audio-video library；Count n sound
There are the numbers of the audio-video of similar audio-video in video；Whether detection number reaches first threshold, and first threshold is greater than 0 and small
In equal to n；If not up to first threshold, by album information typing to audio-video library；Only when n audio-video is in audio-video library
It is middle there are the number of the audio-video of similar audio-video be less than first threshold when, just by album information typing to audio-video library；It solves
When the album of same names is not present in backstage manager in finding audio repository, by the song in the album name and album
Qu Jinhang typing leads to the different albums that may there are problems that same song in audio repository；Reach only when n sound view
When frequency is less than first threshold there are the number of the audio-video of similar audio-video in audio-video library, album information is just entered into sound
In video library, the different albums that there are problems that same song in audio-video library are avoided.
It should be understood that the above general description and the following detailed description are merely exemplary, this can not be limited
It is open.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is the method flow diagram of album input method provided by one embodiment of the present invention；
Fig. 2 is the method flow diagram for the album input method that another embodiment of the present invention provides；
Fig. 3 A is the flow chart of the sub-step of step 204 in Fig. 2 embodiment provided by one embodiment of the present invention；
Fig. 3 B is the method flow diagram of singer's input method provided by one embodiment of the present invention；
Fig. 4 is the structural block diagram of album input device provided by one embodiment of the present invention；
Fig. 5 is the structural block diagram for the album input device that another embodiment of the present invention provides.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.
Fig. 1 is the method flow diagram of album input method provided by one embodiment of the present invention.This method comprises:
Step 101, album information is obtained, album information includes n audio-video in target album, and n is positive integer.
Step 102, for each audio-video in n audio-video, detecting in audio-video library whether there is and audio-video phase
As similar audio-video.
Step 103, there are the numbers of the audio-video of similar audio-video in n audio-video of statistics.
Step 104, whether detection number reaches first threshold, and first threshold is greater than 0 and is less than or equal to n.
Step 105, if not up to first threshold, by album information typing to audio-video library.
In conclusion album input method provided in this embodiment, by obtaining album information, album information includes target
N audio-video in album, n are positive integer；For each audio-video in n audio-video, detects and whether deposited in audio-video library
In similar audio-video similar to audio-video；There are the numbers of the audio-video of similar audio-video in n audio-video of statistics；Detection
Whether number reaches first threshold, and first threshold is greater than 0 and is less than or equal to n；If not up to first threshold, by album information typing
To audio-video library；Only when there are the numbers of the audio-video of similar audio-video less than the first threshold in audio-video library for n audio-video
When value, just by album information typing to audio-video library；Solving backstage manager, there is no mutually of the same name in finding audio repository
When the album of title, the song in the album name and album is subjected to typing, leads to there may be identical song in audio repository
The problem of different albums of song；Reach only when there are the audio-videos of similar audio-video in audio-video library for n audio-video
When number is less than first threshold, just album information is entered into audio-video library, avoids in audio-video library that there are same songs
Different albums the problem of.
Fig. 2 is the method flow diagram for the album input method that another embodiment of the present invention provides.This method comprises:
Step 201, album information is obtained, album information includes n audio-video in target album, and n is positive integer.
The n audio-video before album typing, first in acquisition target album.
Optionally, the target album in the embodiment of the present invention may include album of songs or video album.Wherein, work as target
When album is album of songs, n audio-video may include the n song in album of songs.Such as: target album is album of songs
" I am extremely busy ", then n audio-video be are as follows: " cowboy is extremely busy ", " sweet tea sweet tea ", " sunlight geek ", " rainbow " and " I is unworthy of " 5 is first
Song.When target album is video album, n audio-video is n video in video album.Such as: target album is video
Album " brother of running ", then n video is the video of updated 16 phase program.
Illustratively, the target album got is album of songs " I am extremely busy ", and n audio-video in target album is 5
Different songs, then the corresponding relationship between the target album got and n audio-video, as shown in Table 1:
Table one
Optionally, album information is that backstage manager gets from each audio-video platform, each special getting
After collecting information, the repetition album information got in the same audio-video platform is excluded according to album information first.Such as: backstage
Administrative staff get 100 album informations from my cruel music, firstly, to 100 albums got from my cruel music
Information carries out repeating detection.Optionally, it is detected in 100 album informations according to album name with the presence or absence of identical album name
Album information, it is identical if it exists, then retain an album information in identical album name.Similarly, to from each audio-video
The album information that platform is got all carries out repeating detection respectively, excludes the repetition album got in the same audio-video platform
Information.
Step 202, the audio-video classification in audio-video library is obtained, each audio-video in each audio-video classification is sound view
The similarity of frequency fingerprint is higher than the similar audio-video of second threshold.
Wherein, audio/video fingerprint includes two parts of audio-frequency fingerprint or video finger print.Optionally, audio-frequency fingerprint includes basis
The feature that the combined factors such as melody, entry, tone, tone color and the speed of audio are extracted.Video finger print includes the platform according to video
Word dubs the feature extracted with combined factors such as word speeds.
Such as: include n song in the target album got, then extracts the corresponding audio-frequency fingerprint of every song respectively.
For another example: including n video file in the target album got, then extract the corresponding video of each video file respectively and refer to
Line.
Each audio-video in audio-video classification is that the similarity of audio/video fingerprint is higher than the similar audio-video of second threshold.
Wherein, second threshold can be 80%, 90% etc..
Each audio-video in audio-video library is obtained in advance, and extracts the audio/video fingerprint of each audio-video, according to each
Each audio-video that similarity is higher than second threshold is divided into one kind by the audio/video fingerprint of audio-video.Such as: it is assumed that the second threshold
Value is 95%, has 100 songs in audio-video library, obtains the audio-frequency fingerprint of every song respectively, it is assumed that 100 audio-frequency fingerprints
In there is the similarity between 20 audio-frequency fingerprints to be all higher than 95%, then using corresponding 20 song of 20 audio-frequency fingerprints as one
A audio classification.
Optionally, classify using the identical each audio-video of audio/video fingerprint in audio-video library as an audio-video.
Such as: there are 100 songs in audio-video library, obtain the audio-frequency fingerprint of every song respectively, it is assumed that has in 100 audio-frequency fingerprints
The value of 20 audio-frequency fingerprints is a, and the value of 30 audio-frequency fingerprints is b, and the value of 50 audio-frequency fingerprints is c；Then by the value of audio-frequency fingerprint
For a 20 songs as an audio classification；Using 30 songs that the value of audio-frequency fingerprint is b as another audio classification；
Using 50 songs that the value of audio-frequency fingerprint is c as another audio classification.
Step 203, for each audio-video in n audio-video, the audio/video fingerprint of audio-video is obtained.
After getting n audio-video in target album, the audio-video for extracting each audio-video in n audio-video refers to
Line.
For obtaining the audio/video fingerprint of each audio-video in n audio-video and obtaining audio-video in the embodiment of the present invention
The sequencing of audio-video classification in library is not especially limited.
It step 204, will be in audio/video fingerprint and the classification of each audio-video for each audio-video in n audio-video
The audio/video fingerprint of part audio-video be compared, detect in audio-video library and regarded with the presence or absence of similar sound similar to audio-video
Frequently.
Optionally, step 204 may include following sub-step, as shown in Figure 3A:
Step 204a classifies for i-th of audio-video in the classification of each audio-video, calculates audio/video fingerprint and part sound
Similarity between the audio/video fingerprint of each audio-video in video, 1≤i≤N, N are the audio-video classification in audio-video library
Total number.
For each audio-video in n audio-video, the audio/video fingerprint of each audio-video is obtained.For each audio-video
I-th of audio-video classification in classification calculates the part sound in the audio/video fingerprint and i-th of audio-video classification of each audio-video
Similarity between the audio/video fingerprint of video.
Optionally, each audio-video classification that will acquire carries out the sequence of random order, classifies from first audio-video
Start, calculates the phase between the audio/video fingerprint of each audio-video and the audio/video fingerprint of the part audio-video in audio-video classification
Like degree.
Optionally, the audio-video of the audio/video fingerprint and the part audio-video in audio-video classification that calculate each audio-video refers to
Similarity between line may include: for each audio-video in n audio-video, by the audio/video fingerprint of each audio-video with
Any one audio/video fingerprint in each audio-video classification is compared, and the audio/video fingerprint and sound for calculating each audio-video regard
Similarity between the audio/video fingerprint of part audio-video in frequency division class.
Optionally, the audio-video of the audio/video fingerprint and the part audio-video in audio-video classification that calculate each audio-video refers to
Similarity between line can also include: the value by the audio/video fingerprint in the classification of each audio-video according to sequence from big to small
Arrangement obtains three audio/video fingerprints that audio/video fingerprint in each audio-video classification is maximum value, minimum value and median, right
Each audio-video in n audio-video, will be in the audio/video fingerprint of each audio-video and each audio-video got classification
Three audio/video fingerprints be compared, calculate each audio-video audio/video fingerprint and audio-video classification in part audio-video
Audio/video fingerprint between similarity.
Step 204b, if there is the similarity more than second threshold in each similarity being calculated, it is determined that sound view
There is similar audio-video similar to audio-video in frequency library.
It is deposited when in each similarity that the audio/video fingerprint with the part audio-video in the i-th assonance visual classification is calculated
It is being more than the similarity of second threshold, is illustrating there is similar audio-video similar to the audio-video in the i-th assonance visual classification, then
It determines and is present in the similar similar audio-video of audio-video in audio-video library.
Such as: the audio/video fingerprint of each audio-video in n audio-video and first kind audio-video classify in it is any one
There is the similarity more than second threshold in the n similarity that a audio/video fingerprint is calculated, then illustrates first kind audio-video
There is similar audio-video similar to the audio-video in n audio-video in classification.For another example: for one in n audio-video
Audio-video, calculate and the second assonance visual classification in three audio/video fingerprints between similarity, obtain three similarities, if
In three similarities there are at least one be more than second threshold, then illustrate in the second assonance visual classification exist and n audio-video
In the similar similar audio-video of audio-video.
Step 204c, if there is no the similarities more than second threshold in each similarity being calculated, in i < N
When, i=i+1 is enabled, executes step 204a again.
When in each similarity that the audio/video fingerprint with the part audio-video in the i-th assonance visual classification is calculated not
In the presence of the similarity for being more than second threshold, illustrate that there is no similar sounds similar to the audio-video to regard in the i-th assonance visual classification
Frequently, then i=i+1 is enabled, step 204a is continued to execute.
Such as: for each audio-video in n audio-video, the audio/video fingerprint and first of each audio-video is calculated first
Similarity between the audio/video fingerprint of part audio-video in assonance visual classification is obtained when with first kind audio-video classified calculating
To similarity be no greater than second threshold when, then calculate and the audio-video of the part audio-video in the second assonance visual classification refer to
Similarity between line is then calculated when the similarity being calculated with the second assonance visual classification is no greater than second threshold
Similarity between the audio/video fingerprint of the part audio-video in third assonance visual classification, and so on, until calculating
Similarity be higher than second threshold or i > N until.
For each audio-video in n audio-video, regarded by the audio/video fingerprint and i-th of sound that calculate each audio-video
Similarity between the audio/video fingerprint of each audio-video in the middle part of frequency division class in partial video is classified when in i-th of audio-video
When middle appearance similar audio-video similar to the audio-video, do not need calculate with other audio-videos classification in audio/video fingerprint it
Between similarity, reduce calculation amount, save the resource of server.
Step 205, there are the numbers of the audio-video of similar audio-video in n audio-video of statistics.
If for each audio-video in n audio-video, in the audio/video fingerprint of n audio-video and the classification of each audio-video
Part audio-video audio/video fingerprint similarity in exist higher than second threshold similarity, illustrate exist in audio-video library
Similar audio-video similar to the audio-video in n audio-video, then count in n audio-video to there are similar sounds in audio-video library
The number of the audio-video of video.
Optionally, for each audio-video in n audio-video, all there may be the sounds with the audio-video in audio-video library
The similarity of video finger print is higher than the similar audio-video of second threshold；Therefore for n audio-video, exist in audio-video library and n
The similar similar audio-video of part audio-video in a audio-video, alternatively, existing and the whole in n audio-video in audio-video library
The similar similar audio-video of audio-video.Such as: it include 5 different songs in target album, each sound in audio-video library
The similarity that audio/video fingerprint corresponding with every song can be found in visual classification is higher than the similar audio-video of second threshold,
It is then 5 there are the number of similar songs in 5 songs of statistics.For another example: including 5 different songs, In in target album
The similarity of audio/video fingerprint corresponding with wherein 3 songs is only able to find in each audio-video classification in audio-video library higher than the
The similar audio-video of two threshold values is then 3 there are the number of similar songs in 5 songs of statistics.
Step 206, whether detection number reaches first threshold, and first threshold is greater than 0 and is less than or equal to n.
By there are the numbers of the audio-video of similar audio-video to be compared with first threshold in the n audio-video counted on,
Judge that there are the numbers of the audio-video of similar audio-video whether to reach first threshold in n audio-video.
It will appear two kinds of testing results for the detection process in step 206, when testing result is to reach first threshold,
Then follow the steps 207；When testing result is not up to first threshold, 208 are thened follow the steps.
It step 207, will if there are the numbers of the audio-video of similar audio-video to be not up to first threshold in n audio-video
Album information typing is to audio-video library.
When being not up to first threshold there are the number of the audio-video of similar audio-video in the n audio-video counted on, say
In bright audio-video library and there is no to the similar similar audio-video of each audio-video in n audio-video in target album, then general
Album information typing is into audio-video library.
Step 208, it if there are the numbers of the audio-video of similar audio-video to reach first threshold in n audio-video, detects
It whether there is the album of the same name with album name in audio-video library.
Wherein, album information further includes the album name of target album.
When reaching first threshold there are the number of the audio-video of similar audio-video in the n audio-video counted on, explanation
There is similar audio-video similar to the part audio-video in n audio-video in audio-video library, then detection is in audio-video library
It is no to there is album identical with the album name of target album.
Optionally, the corresponding album name of each album in audio-video library is obtained, by the album name and sound of target album
The corresponding album name of each album is compared in video library, detects in the corresponding album name of each album in audio-video library
With the presence or absence of album identical with the album name of target album.
Optionally, if there are the numbers of the audio-video of similar audio-video to reach first threshold in the n audio-video counted on
When, it can directly execute step 210.
Step 209, if there is the album of the same name with album name in audio-video library, it is determined that target album belongs to audio-video
It is middle to detect obtained album.
If being compared by the album name of target album album name corresponding with album each in audio-video library
Cheng Zhong has found that the album name of target album is identical as the corresponding album name of some album in audio-video library, then target is special
Volume it is determined as the association album of album of the same name in audio-video library.Namely think that target album and album of the same name in audio-video library are phase
Same album, therefore album information is not entered into audio-video library.
Optionally, target album is determined as the association album of album of the same name in audio-video library includes: in audio-video library
It includes n audio-video that mark, which exists in association album and association album, in album of the same name.
Such as: in target album " I am extremely busy " corresponding 5 song be " cowboy is extremely busy ", " sweet tea sweet tea ", " sunlight geek ",
" I is unworthy of " and " rainbow "；And the album name of target album " I am extremely busy " " I am extremely busy " corresponding with the album in audio-video library
Identical, then in album " I am extremely busy " there is association album and mark the 5 first songs for including in association album in mark in audio-video library
Song is " cowboy is extremely busy ", " sweet tea sweet tea ", " sunlight geek ", " I is unworthy of " and " rainbow ".
Step 210, the album of the same name with album name if it does not exist, then detect of the similar audio-video in same album
Whether number reaches first threshold.
It may include following two ways that whether the number for detecting the similar audio-video in same album, which reaches first threshold:
As the first possible implementation, when the album of the same name with album name is not present in audio-video library, then
Audio-video library each similar audio-video similar to the audio-video in n audio-video is obtained, after getting each similar audio-video,
Whether the number for detecting the similar audio-video in same album reaches first threshold.
Such as: in target album " I am extremely busy " corresponding 5 song be " cowboy is extremely busy ", " sweet tea sweet tea ", " sunlight geek ",
" I is unworthy of " and " rainbow ".With " cowboy is extremely busy " there is the song of similar audio-frequency fingerprint there are 3 songs in audio-video library, with
" sweet tea sweet tea " there is the song of similar audio-frequency fingerprint to have 3 songs, have the song of similar audio-frequency fingerprint with " sunlight geek "
Song has 4 head, with " I is unworthy of " there is the song of similar audio-frequency fingerprint to have 4, has similar audio-frequency fingerprint with " rainbow "
Song has 5, then detection and " cowboy is extremely busy " similar 3 song and " sweet tea sweet tea " similar 3 song and " sunlight residence
Belong in similar 4 song of male " and " I is unworthy of " similar 4 song and 5 song similar with " rainbow " same special
Whether the number for the similar audio-video collected reaches first threshold.
As second of possible implementation, the audio-video classification in audio-video library in each album is obtained.Detection is each
In audio-video classification in a album with the presence or absence of with the similarity of the audio/video fingerprint of the audio-video in n audio-video higher than the
The similar audio-video of two threshold values.That is, detecting the audio/video fingerprint and the sound in same album of each audio-video in n audio-video
Whether the number that the similarity between the audio/video fingerprint of visual classification is higher than second threshold reaches first threshold.
Such as: target album includes 5 different songs in " I am extremely busy ".Wherein, " cowboy is extremely busy " corresponding audio
Fingerprint is a, and " sweet tea sweet tea " corresponding audio-frequency fingerprint is b, and " sunlight geek " corresponding audio-frequency fingerprint is c, and " I is unworthy of " is corresponding
Audio-frequency fingerprint is d, and " rainbow " corresponding audio-frequency fingerprint is e；Detect in same album audio-video classification in exist respectively with sound
Similarity between frequency fingerprint a, audio-frequency fingerprint b, audio-frequency fingerprint c, audio-frequency fingerprint d and audio-frequency fingerprint e is higher than the phase of second threshold
Whether reach first threshold like the number of audio-video.
In the second possible implementation, when the audio/video fingerprint number in the same album in audio-video library is greater than
When the audio/video fingerprint number of n audio-video, detects in same album and whether regarded comprising similar sound similar to n audio-video
Frequently；When the audio/video fingerprint number of the same album in audio-video library is equal to the audio/video fingerprint number of n audio-video, detection
Whether to n audio-video there is one-to-one similar audio-video in the audio-video of same album.
It will appear two kinds of testing results for the detection process in step 210, when testing result is to reach first threshold,
Then follow the steps 211；When testing result is not up to first threshold, 212 are thened follow the steps.
Step 211, if the number of the similar audio-video in same album reaches first threshold, it is determined that target album belongs to
Detect obtained album.
If the number of the similar audio-video in same album reaches first threshold, illustrate to exist in audio-video library special with target
The same or similar album is collected, then target album is determined to belong to the album detected in audio-video library.Namely think mesh
The album detected in mark album and audio-video library is that there are associated albums, therefore album information are not entered into audio-video
In library.
It optionally, include: to be regarded in sound by the association album that target album is determined as the album detected in audio-video library
It includes n audio-video that mark, which exists in association album and association album, in the album detected in frequency library.
Step 212, if the number of the similar audio-video in same album is not up to first threshold, by album information typing
To audio-video library.
If the number of the similar audio-video in same album is not up to first threshold, illustrate to be not present in audio-video library and mesh
The same or similar album of album is marked, then by album information typing to audio-video library.
It optionally, include by the n sound view in album name and target album into audio-video library by album information typing
Frequency typing is into audio-video library.
In conclusion album input method provided in this embodiment, by obtaining album information, album information includes target
N audio-video in album, n are positive integer；For each audio-video in n audio-video, detects and whether deposited in audio-video library
In similar audio-video similar to audio-video；There are the numbers of the audio-video of similar audio-video in n audio-video of statistics；Detection
Whether number reaches first threshold, and first threshold is greater than 0 and is less than or equal to n；If not up to first threshold, by album information typing
To audio-video library；Only when there are the numbers of the audio-video of similar audio-video less than the first threshold in audio-video library for n audio-video
When value, just by album information typing to audio-video library；Solving backstage manager, there is no mutually of the same name in finding audio repository
When the album of title, the song in the album name and album is subjected to typing, leads to there may be identical song in audio repository
The problem of different albums of song；Reach only when there are the audio-videos of similar audio-video in audio-video library for n audio-video
When number is less than first threshold, just album information is entered into audio-video library, avoids in audio-video library that there are same songs
Different albums the problem of.
In addition, passing through the part in the audio/video fingerprint of each audio-video in n audio-video and the classification of each audio-video
The audio/video fingerprint of audio-video is compared so that only need to by the part audio/video fingerprint in audio/video fingerprint and audio-video library into
Row compares, and reduces calculation amount, saves server resource.
Moreover, can be regarded by whether there is the album of the same name with album name in detection audio-video library to avoid typing sound
Already present identical album or similar album in frequency library.
It should be noted is that in the embodiment of the present invention, when the part in the audio/video fingerprint of n audio-video is present in
It is complete for album when in the audio-video classification of same album, by the n audio-video whole in album name and target album
Typing is to audio-video library.In actual operation, as a kind of possible implementation, when in the audio/video fingerprint of n audio-video
Part when being present in the audio-video classification of same album, can n audio-video in typing album name and target album
Audio/video fingerprint be not included in same album audio-video classification in corresponding audio-video.
Such as: the audio-frequency fingerprint of the album of songs in audio-video library includes: audio-frequency fingerprint a, audio-frequency fingerprint c, audio-frequency fingerprint
E, five audio/video fingerprints of audio-frequency fingerprint f and audio-frequency fingerprint g, and target album includes 5 different songs in " I am extremely busy ".
Wherein, " cowboy is extremely busy " corresponding audio-frequency fingerprint is a, and " sweet tea sweet tea " corresponding audio-frequency fingerprint is b, " sunlight geek " corresponding sound
Frequency fingerprint is c, and " I is unworthy of " corresponding audio-frequency fingerprint is d, and " rainbow " corresponding audio-frequency fingerprint is e.Therefore, " cowboy is extremely busy ",
" sunlight geek " and " rainbow " corresponding audio-frequency fingerprint are present in the album of songs of audio-video library, " sweet tea sweet tea " and " I not
With " corresponding audio-frequency fingerprint is not present in the album of audio-video library；It is then used as a kind of possible implementation, considers album
Integrality, by album name " I am extremely busy " and all typings of corresponding 5 song into audio-video library；As alternatively possible
Implementation considers the repeatability of song, extremely by album name " I am extremely busy " and corresponding " sweet tea sweet tea " and " I is unworthy of " typing
In audio-video library.
It needs to illustrate on the other hand, in embodiment shown in Fig. 2, detecting in step 208 whether there is in audio-video library
Whether reach the first threshold to the number for detecting the similar audio-video in same album in album name album of the same name and step 210
Value is optional detecting step, and therefore, in embodiment shown in Fig. 2, step 208 to step 212 is optional execution step, In
In practical implementation, it can not be performed simultaneously.
Based in embodiment shown in Fig. 2, for album name and corresponding n audio-video.When Fig. 2 reality
Applying the target album in example is album of songs, when n audio-video is n song, further includes having and mesh during being actually typing
Mark the typing of the corresponding singer of album.Specific embodiment is as shown in Figure 3B.
Step 301, target singer corresponding to target album is obtained.
After getting album information, the target album in album information is read, target album is obtained according to target album
Corresponding target singer.
Step 302, for every song in n song, song in audio-video library is obtained according to the song fingerprints of song
Singer corresponding to every song in affiliated categorizing songs.
After getting album information, the n song in target album is read, extracts the corresponding song fingerprints of every song,
The similarity in audio-video library between the song fingerprints, which is obtained, according to song fingerprints is higher than predetermined threshold or identical song
Categorizing songs where Qu Zhiwen after getting the categorizing songs in audio-video library, obtain every song institute in the categorizing songs
Corresponding singer.
A categorizing songs or multiple songs point may be got in audio-video library according to the song fingerprints of n song
Class obtains the corresponding singer of every song in each categorizing songs respectively.
Such as: in target album " I am extremely busy " corresponding 5 song be " cowboy is extremely busy ", " sweet tea sweet tea ", " sunlight geek ",
" I is unworthy of " and " rainbow ", wherein " cowboy is extremely busy " corresponding audio-frequency fingerprint is a, and " sweet tea sweet tea " corresponding audio-frequency fingerprint is b,
" sunlight geek " corresponding audio-frequency fingerprint is c, and " I is unworthy of " corresponding audio-frequency fingerprint is d, and " rainbow " corresponding audio-frequency fingerprint is
E, then be directed to target album, respectively obtain audio-video library sound intermediate frequency fingerprint a, audio-frequency fingerprint b, audio-frequency fingerprint c, audio-frequency fingerprint d and
Singer corresponding to every song in the affiliated categorizing songs of audio-frequency fingerprint e.
Step 303, whether detection target singer is a member in each singer got.
Corresponding to every song in getting target singer and audio-video library in categorizing songs corresponding with n song
Singer after, target singer is compared with each singer got, detection target singer whether be get it is each
A member in singer.
In the detection process of step 303, it may appear that two different testing results, when target singer be get it is each
When a member in a singer, step 304 is executed；When target singer is not a member in each singer got, step is executed
Rapid 305.
Step 304, if testing result is yes, it is determined that target singer have existed in audio-video library.
When there is the singer of the same name with target singer in each singer got, then target singer is determined as audio-video
The association singer of the singer detected in library.Namely think that the singer detected in target singer and audio-video library is same
A singer, therefore target singer is not entered into audio-video library.
It optionally, include: to be regarded in sound by the association singer that target singer is determined as the singer detected in audio-video library
It is the corresponding album of target singer is mesh that mark, which has association singer and mark association singer, in the singer detected in frequency library
Mark album.
Optionally, if detect that target singer is a member in each singer got, each song can also be detected
Whether the singer number of the same name with target singer is 1 in hand.Wherein, including following detecting step:
The first step, when target singer exists with each singer for getting, detecting in each singer got is
No existence anduniquess and target singer singer of the same name.
Second step, when singer's number of the same name with target singer in the first step be 1, it is determined that target singer have existed with
Audio-video library.
Third step then detects each singer got when singer's number of the same name with target singer in the first step is not 1
In whether include target singer.
Such as: target singer is Xiao Ming, and there are Zhang little Ming in the singer got, then illustrate include in the singer got
Target singer.
Target singer is not entered into then by the 4th step when including target singer in each singer got in third step
In audio-video library, or, whether artificial judgment is entered into target singer in audio-video library.
5th step, when in each singer got in third step do not include target singer, then detect get it is each
Whether there are of the same name with target singer by singer.
6th step, when there are of the same name, then target singer being determined as the singer detected in audio-video library in the 5th step
Association singer.
It is of the same name if it exists, then target singer is not entered into audio-video library, or, whether artificial judgment records target singer
Enter into audio-video library.
7th step, it is of the same name when being not present in the 5th step, then follow the steps 305.
Step 305, if testing result be it is no, according to the song fingerprints of song obtain audio-video library in other categorizing songs
In each song corresponding to singer.
If target singer is not present in each singer got, illustrate that target singer is not present in audio-video library
In categorizing songs corresponding with the song fingerprints of n song in singer corresponding to every song.Therefore, it obtains in audio-video library
Singer corresponding to every song in other categorizing songs.
Step 306, detection target singer whether be in singer corresponding to each song in other categorizing songs one
Member.
It, will after singer corresponding to every song in getting target singer and audio-video library in other categorizing songs
Target singer is compared with singer corresponding to every song in the other categorizing songs got, and detection target singer is
No is a member in singer corresponding to every song in the other categorizing songs got.
Step 307, if testing result be it is no, by target singer typing to audio-video.
If target singer is not a member in the corresponding singer of every song in audio-video record in other categorizing songs,
Illustrate that there is no target singers in audio-video library, therefore, by target singer typing into audio-video library.
Step 308, if testing result is yes, it is determined that target singer has existed and audio-video library.
If target singer is a member in the corresponding singer of every song in audio-video record in other categorizing songs, say
There may be target singers in bright audio-video library, therefore, in order to avoid the repetition typing of singer, determine that target singer has existed
In audio-video library, not by target singer typing into audio-video library.
Optionally, if target singer is one in the corresponding singer of every song in audio-video record in other categorizing songs
Member, then illustrate that there may be target singers in audio-video library, and therefore, whether artificial judgment by target singer is entered into audio-video library
In.
It should be noted is that step 301 to step 308 album typing step shown in Fig. 2 embodiment it
Preceding execution can also execute after album typing step.Namely it can typing again after first typing singer in the embodiment of the present invention
Album information；It can also the corresponding singer of typing again after first typing album information.The embodiment of the present invention is to album typing and singer
The sequencing of typing is not especially limited.
Fig. 4 is the structural block diagram of album input device provided by one embodiment of the present invention.The album input device packet
It includes:
Album obtains module 410, and for obtaining album information, album information includes n audio-video in target album, n
For positive integer.
Approx imately-detecting module 420, for detecting and whether being deposited in audio-video library for each audio-video in n audio-video
In similar audio-video similar to audio-video.
Number statistical module 430, for counting, there are the numbers of the audio-video of similar audio-video in n audio-video.
Number detection module 440, for detecting whether number reaches first threshold, first threshold is greater than 0 and is less than or equal to
n。
Album recording module 450 is used in not up to first threshold, by album information typing to audio-video library.
In conclusion album input device provided in this embodiment, by obtaining album information, album information includes target
N audio-video in album, n are positive integer；For each audio-video in n audio-video, detects and whether deposited in audio-video library
In similar audio-video similar to audio-video；There are the numbers of the audio-video of similar audio-video in n audio-video of statistics；Detection
Whether number reaches first threshold, and first threshold is greater than 0 and is less than or equal to n；If not up to first threshold, by album information typing
To audio-video library；Only when there are the numbers of the audio-video of similar audio-video less than the first threshold in audio-video library for n audio-video
When value, just by album information typing to audio-video library；Solving backstage manager, there is no mutually of the same name in finding audio repository
When the album of title, the song in the album name and album is subjected to typing, leads to there may be identical song in audio repository
The problem of different albums of song；Reach only when there are the audio-videos of similar audio-video in audio-video library for n audio-video
When number is less than first threshold, just album information is entered into audio-video library, avoids in audio-video library that there are same songs
Different albums the problem of.
Fig. 5 is the structural block diagram for the album input device that another embodiment of the present invention provides.The album input device
Include:
First obtains module 510, for obtaining target singer corresponding to target album.
Optionally, target album is album of songs, and n audio-video is n song.
Second obtains module 520, for obtaining sound according to the song fingerprints of song for every song in n song
Singer corresponding to every song in video library in the affiliated categorizing songs of song.
First detection module 530, for detecting whether target singer is a member in each singer got.
Singer's determining module 540, for determining that target singer has existed and audio-video library when testing result, which is, is.
Third obtains module 550, for being obtained in audio-video library according to the song fingerprints of song when testing result is no
Singer corresponding to each song in other categorizing songs.
Second detection module 560, for detecting whether target singer is corresponding to each song in other categorizing songs
Singer in a member.
Singer's recording module 570 is used for when testing result is no, by target singer typing to audio-video.
Album obtains module 580, and for obtaining album information, album information includes n audio-video in target album, n
For positive integer.
Approx imately-detecting module 590, for detecting and whether being deposited in audio-video library for each audio-video in n audio-video
In similar audio-video similar to audio-video.
Optionally, approx imately-detecting module 590 may include: classification acquisition submodule 591,592 and of fingerprint acquisition submodule
Fingerprint comparison submodule 593.
Classification acquisition submodule 591, it is each in each audio-video classification for obtaining the classification of the audio-video in audio-video library
A audio-video is that the similarity of audio/video fingerprint is higher than the similar audio-video of second threshold.
Fingerprint acquisition submodule 592, for obtaining the audio-video of audio-video for each audio-video in n audio-video
Fingerprint.
Fingerprint comparison submodule 593, for for each audio-video in n audio-video, by audio/video fingerprint and each
The audio/video fingerprint of part audio-video in the classification of a audio-video is compared, and detect whether there is and audio-video in audio-video library
Similar similar audio-video.
Optionally, fingerprint comparison submodule 593 may include: similar computing unit 593a and similar determination unit 593b.
Similar computing unit 593a calculates audio-video for classifying for i-th of audio-video in the classification of each audio-video
Similarity between the audio/video fingerprint of each audio-video in fingerprint and part audio-video, 1≤i≤N, N are in audio-video library
Audio-video classification total number.
Similar determination unit 593b, for there is the similarity more than second threshold in each similarity being calculated
When, determine there is similar audio-video similar to audio-video in audio-video library.
Similar computing unit 593a is also used to that the phase more than second threshold is not present in each similarity being calculated
When seemingly spending, in i < N, i=i+1 is enabled, executes the sound for calculating each audio-video in audio/video fingerprint and part audio-video again
The step of similarity between video finger print.
Number statistical module 610, for counting, there are the numbers of the audio-video of similar audio-video in n audio-video.
Number detection module 620, for detecting whether number reaches first threshold, first threshold is greater than 0 and is less than or equal to
n。
Optionally, album information further includes the album name of target album；The device, further includes:
Title detection module 630, for detecting in audio-video library with the presence or absence of the album of the same name with album name；
Album detection module 640, in the absence of testing result is, executing the similar sound view detected in same album
The step of whether number of frequency reaches first threshold.
Optionally, the device, further includes:
Album detection module 640, for when reaching first threshold, detecting the number of the similar audio-video in same album
Whether first threshold is reached；
Album determining module 650, for when reaching first threshold, determining that target album belongs to the album that detection obtains.
Album recording module 660 is used in not up to first threshold, by album information typing to audio-video library.
In conclusion album input device provided in this embodiment, by obtaining album information, album information includes target
N audio-video in album, n are positive integer；For each audio-video in n audio-video, detects and whether deposited in audio-video library
In similar audio-video similar to audio-video；There are the numbers of the audio-video of similar audio-video in n audio-video of statistics；Detection
Whether number reaches first threshold, and first threshold is greater than 0 and is less than or equal to n；If not up to first threshold, by album information typing
To audio-video library；Only when there are the numbers of the audio-video of similar audio-video less than the first threshold in audio-video library for n audio-video
When value, just by album information typing to audio-video library；Solving backstage manager, there is no mutually of the same name in finding audio repository
When the album of title, the song in the album name and album is subjected to typing, leads to there may be identical song in audio repository
The problem of different albums of song；Reach only when there are the audio-videos of similar audio-video in audio-video library for n audio-video
When number is less than first threshold, just album information is entered into audio-video library, avoids in audio-video library that there are same songs
Different albums the problem of.
In addition, passing through the part in the audio/video fingerprint of each audio-video in n audio-video and the classification of each audio-video
The audio/video fingerprint of audio-video is compared so that only need to by the part audio/video fingerprint in audio/video fingerprint and audio-video library into
Row compares, and reduces calculation amount, saves server resource.
Moreover, can be regarded by whether there is the album of the same name with album name in detection audio-video library to avoid typing sound
Already present identical album or similar album in frequency library.
It should be understood that the device of album typing provided by the above embodiment is in typing album, only with above-mentioned each function
Can module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different functions
Module is completed, i.e., the internal structure of equipment is divided into different functional modules, described above all or part of to complete
Function.In addition, the device of album typing provided by the above embodiment and the embodiment of the method for album typing belong to same design,
Specific implementation process is detailed in embodiment of the method, and which is not described herein again.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Those of ordinary skill in the art will appreciate that realizing that all or part of the steps of above-described embodiment can pass through hardware
Complete, relevant hardware can also be instructed to complete by program, program can store in a kind of computer-readable storage
In medium, storage medium mentioned above can be read-only memory, disk or CD etc..
The foregoing is merely a prefered embodiment of the invention, is not intended to limit the invention, all in the spirit and principles in the present invention
Within, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention.
Claims (12)
1. a kind of album input method, which is characterized in that the described method includes:
Album information is obtained, the album information includes n audio-video in target album, and n is positive integer；
The audio-video classification in the audio-video library is obtained, each audio-video in each audio-video classification is audio/video fingerprint
Similarity is higher than the similar audio-video of second threshold；
For each audio-video in the n audio-video, the audio/video fingerprint of the audio-video is obtained；
For each audio-video in the n audio-video, by the portion in the audio/video fingerprint and the classification of each audio-video
The audio/video fingerprint of partial video is successively compared, and detects in the audio-video library with the presence or absence of similar with the audio-video
Similar audio-video, when detect in the part audio-video in the classification of each audio frequency and video exist it is similar similar to the audio-video
When audio-video, determine there is similar audio-video similar to the audio-video in the audio-video library；
Count in the n audio-video that there are the numbers of the audio-video of similar audio-video；
Detect whether the number reaches first threshold, the first threshold is greater than 0 and is less than or equal to n；
If the not up to described first threshold, by the album information typing to the audio-video library.
2. the method according to claim 1, wherein each audio-video in the n audio-video,
The audio/video fingerprint of part audio-video in the audio/video fingerprint and the classification of each audio-video is successively compared, is detected
It whether there is similar audio-video similar to the audio-video in the audio-video library, comprising:
For i-th of audio-video classification in each audio-video classification, the audio/video fingerprint and the part sound are calculated
Similarity between the audio/video fingerprint of each audio-video in video, 1≤i≤N, N are the audio-video in the audio-video library
The total number of classification；
If there is the similarity more than the second threshold in each similarity being calculated, it is determined that the audio-video
There is similar audio-video similar to the audio-video in library；
If there is no the similarities more than the second threshold to enable i in i < N in each similarity being calculated
=i+1, the audio-video for executing the calculating audio/video fingerprint and each audio-video in the part audio-video again refer to
The step of similarity between line.
3. the method according to claim 1, wherein the method, further includes:
If reaching the first threshold, whether the number for detecting the similar audio-video in same album reaches first threshold
Value；
If reaching the first threshold, it is determined that the target album belongs to the album that detection obtains.
4. according to the method described in claim 3, it is characterized in that, the album information further includes the album of the target album
Title；The method also includes:
It detects in the audio-video library with the presence or absence of the album of the same name with the album name；
If testing result is that there is no whether the number for executing the similar audio-video in the same album of detection reaches described
The step of first threshold.
5. method according to any one of claims 1 to 4, which is characterized in that the target album is album of songs, the n
A audio-video is n song；The method also includes:
Obtain target singer corresponding to the target album；
For every song in the n song, obtained described in the audio-video library according to the song fingerprints of the song
Singer corresponding to every song in the affiliated categorizing songs of song；
Detect whether the target singer is a member in each singer got；
If testing result is yes, it is determined that the target singer has existed and the audio-video library.
6. according to the method described in claim 5, it is characterized in that, the method also includes:
If testing result be it is no, obtained in the audio-video library in other categorizing songs according to the song fingerprints of the song
Singer corresponding to each song；
Detect whether the target singer is a member in singer corresponding to each song in other categorizing songs；
If testing result be it is no, by the target singer typing to the audio-video.
7. a kind of album input device, which is characterized in that described device includes:
Album obtains module, and for obtaining album information, the album information includes n audio-video in target album, and n is positive
Integer；
Approx imately-detecting module, for for each audio-video in the n audio-video, detect in audio-video library with the presence or absence of with
The similar similar audio-video of the audio-video；
Number statistical module, for counting, there are the numbers of the audio-video of similar audio-video in the n audio-video；
Number detection module, for detecting whether the number reaches first threshold, the first threshold is greater than 0 and is less than or equal to
n；
Album recording module is used in the not up to first threshold, by the album information typing to the audio-video library；
The approx imately-detecting module, comprising:
Classification acquisition submodule, it is each in each audio-video classification for obtaining the classification of the audio-video in the audio-video library
Audio-video is that the similarity of audio/video fingerprint is higher than the similar audio-video of second threshold；
Fingerprint acquisition submodule, for obtaining the audio-video of the audio-video for each audio-video in the n audio-video
Fingerprint；
Fingerprint comparison submodule, for for each audio-video in the n audio-video, by the audio/video fingerprint and each
The audio/video fingerprint of part audio-video in a audio-video classification is successively compared, and detecting whether there is in the audio-video library
Similar audio-video similar to the audio-video exists and institute when detecting in the part audio-video in each audio frequency and video classification
When stating the similar similar audio-video of audio-video, determine in the audio-video library that there is similar sound similar to the audio-video regards
Frequently.
8. device according to claim 7, which is characterized in that the fingerprint comparison submodule, comprising:
Similar computing unit calculates the audio-video for classifying for i-th of audio-video in each audio-video classification
Similarity between fingerprint and the audio/video fingerprint of each audio-video in the part audio-video, 1≤i≤N, N are the sound
The total number of audio-video classification in video library；
Similar determination unit, for there is the similarity more than the second threshold in each similarity being calculated
When, determine there is similar audio-video similar to the audio-video in the audio-video library；
The similar computing unit, being also used to be not present in each similarity being calculated is more than the second threshold
Similarity when, in i < N, enable i=i+1, execute described calculate in the audio/video fingerprint and the part audio-video again
Each audio-video audio/video fingerprint between similarity the step of.
9. device according to claim 7, which is characterized in that described device, further includes:
Album detection module, the number for when reaching the first threshold, detecting the similar audio-video in same album are
It is no to reach the first threshold；
Album determining module, for when reaching the first threshold, determine the target album belong to detection obtain it is described
Album.
10. device according to claim 9, which is characterized in that the album information further includes the special of the target album
Collect title；
Described device, further includes:
Title detection module, for detecting in the audio-video library with the presence or absence of the album of the same name with the album name；
The album detection module, in the absence of testing result is, executing the similar sound in the same album of detection
The step of whether number of video reaches the first threshold.
11. according to any device of claim 7 to 10, which is characterized in that the target album is album of songs, described
N audio-video is n song；
Described device, further includes:
First obtains module, for obtaining target singer corresponding to the target album；
Second obtains module, for obtaining institute according to the song fingerprints of the song for every song in the n song
State singer corresponding to every song in the affiliated categorizing songs of song described in audio-video library；
First detection module, for detecting whether the target singer is a member in each singer got；
Singer's determining module, for determining that the target singer has existed and the audio-video library when testing result, which is, is.
12. device according to claim 11, which is characterized in that described device, further includes:
Third obtains module, for obtaining the audio-video library according to the song fingerprints of the song when testing result is no
In singer corresponding to each song in other categorizing songs；
Second detection module, for detecting whether the target singer is corresponding to each song in other categorizing songs
Singer in a member；
Singer's recording module is used for when testing result is no, by the target singer typing to the audio-video.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610173450.XA CN105893463B (en) | 2016-03-23 | 2016-03-23 | Album input method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610173450.XA CN105893463B (en) | 2016-03-23 | 2016-03-23 | Album input method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105893463A CN105893463A (en) | 2016-08-24 |
| CN105893463B true CN105893463B (en) | 2019-11-05 |
Family
ID=57013746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610173450.XA Active CN105893463B (en) | 2016-03-23 | 2016-03-23 | Album input method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105893463B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113127676A (en) * | 2021-05-17 | 2021-07-16 | 杭州网易云音乐科技有限公司 | Information matching method, system, device, storage medium and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101523343A (en) * | 2006-08-04 | 2009-09-02 | 谷歌公司 | Detecting duplicate and near-duplicate files |
| CN101872415A (en) * | 2010-05-06 | 2010-10-27 | 复旦大学 | Video copying detection method being suitable for IPTV |
| CN103095824A (en) * | 2013-01-09 | 2013-05-08 | 广东一一五科技有限公司 | File uploading control method and system |
| CN103177099A (en) * | 2013-03-20 | 2013-06-26 | 深圳先进技术研究院 | Video comparison method and video comparison system |
| CN103678702A (en) * | 2013-12-30 | 2014-03-26 | 优视科技有限公司 | Video duplicate removal method and device |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8463797B2 (en) * | 2010-07-20 | 2013-06-11 | Barracuda Networks Inc. | Method for measuring similarity of diverse binary objects comprising bit patterns |
-
2016
- 2016-03-23 CN CN201610173450.XA patent/CN105893463B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101523343A (en) * | 2006-08-04 | 2009-09-02 | 谷歌公司 | Detecting duplicate and near-duplicate files |
| CN101872415A (en) * | 2010-05-06 | 2010-10-27 | 复旦大学 | Video copying detection method being suitable for IPTV |
| CN103095824A (en) * | 2013-01-09 | 2013-05-08 | 广东一一五科技有限公司 | File uploading control method and system |
| CN103177099A (en) * | 2013-03-20 | 2013-06-26 | 深圳先进技术研究院 | Video comparison method and video comparison system |
| CN103678702A (en) * | 2013-12-30 | 2014-03-26 | 优视科技有限公司 | Video duplicate removal method and device |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113127676A (en) * | 2021-05-17 | 2021-07-16 | 杭州网易云音乐科技有限公司 | Information matching method, system, device, storage medium and electronic equipment |
| CN113127676B (en) * | 2021-05-17 | 2022-07-01 | 杭州网易云音乐科技有限公司 | Information matching method, system, device, storage medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105893463A (en) | 2016-08-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101247470B (en) | Method realized by computer for detecting scene boundaries in videos | |
| US10134440B2 (en) | Video summarization using audio and visual cues | |
| US8867891B2 (en) | Video concept classification using audio-visual grouplets | |
| US8699852B2 (en) | Video concept classification using video similarity scores | |
| CN103927510B (en) | Pattern recognition device and image-recognizing method | |
| Roma et al. | Recurrence quantification analysis features for environmental sound recognition | |
| JP2004229283A (en) | Method for identifying transition of news presenter in news video | |
| JP2003177778A (en) | Audio excerpts extracting method, audio data excerpts extracting system, audio excerpts extracting system, program, and audio excerpts selecting method | |
| WO2012078702A1 (en) | Video key frame extraction using sparse representation | |
| CN110335611B (en) | Voiceprint recognition algorithm evaluation method based on quality dimension | |
| JP4714710B2 (en) | Automatic tagging device, automatic tagging method, automatic tagging program, and recording medium recording the program | |
| Bano et al. | Discovery and organization of multi-camera user-generated videos of the same event | |
| CN105893463B (en) | Album input method and device | |
| US9213703B1 (en) | Pitch shift and time stretch resistant audio matching | |
| Yang et al. | Lecture video browsing using multimodal information resources | |
| Rapantzikos et al. | An audio-visual saliency model for movie summarization | |
| Maheswari et al. | Sports video classification using multi scale framework and nearest neighbor classifier | |
| JP2010038943A (en) | Sound signal processing device and method | |
| Yi et al. | Adaptive threshold based video shot boundary detection framework | |
| Deshpande et al. | Mugec: Automatic music genre classification | |
| Zanganeh et al. | Automatic weak learners selection for pattern recognition and its application in soccer goal recognition | |
| Darji et al. | A review of video classification techniques | |
| Islam et al. | Content-based music retrieval using beat information | |
| Thiruvengatanadhan et al. | Speech/music classification using SVM | |
| Hemalatha et al. | An empirical model for clustering and classification of instrumental music using machine learning technique |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 510660 Guangzhou City, Guangzhou, Guangdong, Whampoa Avenue, No. 315, self - made 1-17 Applicant after: Guangzhou KuGou Networks Co., Ltd. Address before: 510000 B1, building, No. 16, rhyme Road, Guangzhou, Guangdong, China 13F Applicant before: Guangzhou KuGou Networks Co., Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |