CN105184610A - Real-time mobile advertisement synchronous putting method and device based on audio fingerprints - Google Patents
Real-time mobile advertisement synchronous putting method and device based on audio fingerprints Download PDFInfo
- Publication number
- CN105184610A CN105184610A CN201510557641.1A CN201510557641A CN105184610A CN 105184610 A CN105184610 A CN 105184610A CN 201510557641 A CN201510557641 A CN 201510557641A CN 105184610 A CN105184610 A CN 105184610A
- Authority
- CN
- China
- Prior art keywords
- audio
- frequency fingerprint
- frequency
- advertisement
- advertisement audio
- 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.)
- Pending
Links
Abstract
The invention discloses a real-time mobile advertisement synchronous putting method based on audio fingerprints. The method comprises the following steps: extracting advertisement audio fingerprints of advertisement audio, and constructing an advertisement audio fingerprint database; obtaining a live video stream in real time, and extracting video stream audio fingerprints corresponding to the live video stream; searching for the advertisement audio fingerprints in the advertisement audio fingerprint database; comparing whether the advertisement audio fingerprints match the video stream audio fingerprints, and when the video stream audio fingerprints do not match the advertisement audio fingerprints, continuously searching for the advertisement audio fingerprints in the advertisement audio fingerprint database; and when the video stream audio fingerprints match the advertisement audio fingerprints, recording the matching frequency of the advertisement audio corresponding to the advertisement audio fingerprints, and if the matching frequency exceeds a specific threshold, pushing a mobile advertisement corresponding to the advertisement audio to a client. The invention further discloses a real-time mobile advertisement synchronous putting device based on audio fingerprints.
Description
Technical field
The present invention relates to areas of information technology, particularly relate to a kind of synchronous put-on method of real-time moving advertising based on audio-frequency fingerprint and equipment.
Background technology
The mobile Internet epoch, except traditional tv advertisement, moving advertising is also more and more subject to the attention of advertiser, at present, advertiser generally contacts voluntarily or entrusts advertising agency, buy advertisement position by moving advertising platform (as AdView), and provide target group and the release time section roughly of advertisement, moving advertising platform is responsible for throwing in advertisement according to the requirement of advertiser or advertising agency to the application software of a large amount of its platform of access.Moving advertising platform is at present mainly through collecting mass users data, be users classification according to age of user, occupation, Web vector graphic record etc., accurate input advertisement, then carries out charging by the mode such as CPC (clicking cost), CPM (every thousand people's access costs), CPA (responding cost) at every turn at every turn.
Audio-frequency fingerprint (AudioFingerprinting) technology is a kind of emerging audio frequency identification technique, this technology is by the spectrum information of Algorithm Analysis audio frequency, extract the numerical characteristic that it is called " audio-frequency fingerprint ", and use the numerical characteristic of the known audio frequency of magnanimity to build audio fingerprint database.Afterwards, can for unknown audio fragment, identical algorithm is adopted to extract audio-frequency fingerprint, and the audio-frequency fingerprint that search is similar to it in audio fingerprint database, then infer according to certain strategy the known audio frequency mated the most, think that unknown audio fragment is from this known audio frequency, thus complete the identification to unknown audio fragment.Audio fingerprint techniques has been widely used in, in the identification service of music, film, TV, greatly changing the life style of the mankind, the while of user friendly, improve social efficiency.
Television advertising is compared with moving advertising, still there is the innate advantages such as abundant in content, sound visual impact is strong, be still the main channel that consumer produces purchase intention, although carried out targeted ads input according to user habit, but still deposit problem both ways: 1. cannot know whether user watches television advertising; 2. cannot respond the purchase impulsion of user in time, realize seeing while buy.So traditional television advertising is thrown in and is separated with moving advertising put-on method, be difficult to the same user of joint effect, as shown in Figure 1.
Summary of the invention
For this reason, the present invention proposes a kind of new synchronous put-on method of real-time moving advertising based on audio-frequency fingerprint at least partially that can solve the problem and equipment.
According to an aspect of the present invention, provide a kind of synchronous put-on method of real-time moving advertising based on audio-frequency fingerprint, comprising: the advertisement audio-frequency fingerprint extracting advertisement audio frequency, build advertisement audio-frequency fingerprint storehouse; Real-time Obtaining live video stream, and extract and described corresponding video flowing audio-frequency fingerprint; At described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint; Whether more described advertisement audio-frequency fingerprint and described video flowing audio-frequency fingerprint match, and wherein, when described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint do not match, then continue at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint; When described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint match, record the number of times that matches of the advertisement audio frequency corresponding with advertisement audio-frequency fingerprint, if described in match number of times when exceeding specific threshold, then the moving advertising corresponding with described advertisement audio frequency is pushed to client.
Alternatively, according to method of the present invention, wherein, extract the advertisement audio-frequency fingerprint of advertisement audio frequency, build advertisement audio-frequency fingerprint storehouse and comprise: the advertisement audio frequency extracting described advertisement; Multiple advertisement audio-frequency fingerprint is extracted from described advertisement audio frequency; Extract key value and the value value of described advertisement audio-frequency fingerprint; Set up and have the value value of identical key value and the corresponding relation of described key value, create a Hash table, described Hash table is described advertisement audio-frequency fingerprint storehouse; Preferably, the moving advertising corresponding with described advertisement audio frequency is sent to client to comprise: on party in request's platform, throw in the moving advertising corresponding with described advertisement audio frequency; The browse advertisements of more described moving advertising and described client, wherein, when described moving advertising is relevant to described browse advertisements, is pushed to described client described moving advertising.
Alternatively, according to method of the present invention, wherein, from described advertisement audio frequency, extract multiple advertisement audio-frequency fingerprint to comprise: the advertisement audio file described advertisement audio frequency being preserved into consolidation form; For each advertisement audio file gives unique coding, obtained the spectrogram of each advertisement audio file by Fourier Transform Algorithm, wherein, the transverse axis of described spectrogram is time t, and the longitudinal axis is frequency f; By described spectrogram according to certain transverse and longitudinal axle resolution, divide several row rectangular areas, for each region, calculate the extreme point of the highest position of each region energy as each region, wherein, described extreme point represents with transverse and longitudinal coordinate (t, f); Described advertisement audio-frequency fingerprint is obtained according to the relation between described extreme point (t, f).
Alternatively, according to method of the present invention, wherein, obtain described advertisement audio-frequency fingerprint according to the relation between described extreme point (t, f) to comprise: obtain extreme point (t
m, f
m) and (t
n, f
n), then dT=t
n-t
m; F
n: f
m: dT as the key value of described advertisement audio-frequency fingerprint, t
mwith the value value of coding as described advertisement audio-frequency fingerprint.
Alternatively, according to method of the present invention, wherein, Real-time Obtaining live video stream, and extraction comprises with described corresponding video flowing audio-frequency fingerprint: monitor live video stream, extract the video flowing audio frequency corresponding with described live video stream; Video flowing audio-frequency fingerprint is obtained according to certain time interval according to described video flowing audio frequency; Extract k value and the v value of described video flowing audio-frequency fingerprint.
Alternatively, according to method of the present invention, wherein, obtain video flowing audio-frequency fingerprint according to described video flowing audio frequency according to certain time interval to comprise: described video flowing audio frequency is preserved into identical with described advertisement audio format; Obtained the spectrogram of each video flowing audio frequency by Fourier Transform Algorithm, wherein, the transverse axis of described spectrogram is time t ', and the longitudinal axis is frequency f '; By described spectrogram according to certain transverse and longitudinal axle resolution, divide several row rectangular areas, for each region, calculate the extreme point of the highest position of each region energy as each region, wherein, described extreme point represents with transverse and longitudinal coordinate (t ', f '); Described video flowing audio-frequency fingerprint is obtained according to the relation between described extreme point (t ', f ').
Alternatively, according to method of the present invention, wherein, obtain described video flowing audio-frequency fingerprint according to the relation between described extreme point (t ', f ') to comprise: obtain extreme point (t '
m, f '
m) and (t '
n, f '
n), then dT '=t '
n-t '
m; F '
n: f '
m: dT ' as the k value of described video flowing audio-frequency fingerprint, t '
mas the v value of described video flowing audio-frequency fingerprint.
Alternatively, according to method of the present invention, wherein, be the advertisement audio-frequency fingerprint searching the key value identical with k value in described advertisement audio-frequency fingerprint storehouse at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint.
Alternatively, according to method of the present invention, wherein, whether more described advertisement audio-frequency fingerprint matches with described video flowing audio-frequency fingerprint comprises: the v value that the k value of the value value that the key value of more described advertisement audio-frequency fingerprint is corresponding and described video flowing audio-frequency fingerprint is corresponding, wherein, when v value is identical with value value, the coding of the corresponding advertisement audio frequency of described advertisement audio-frequency fingerprint is counted; Whether the counts of calculation code exceedes specific threshold.
According to another aspect of the present invention, additionally provide a kind of equipment synchronously thrown in based on the real-time moving advertising of audio-frequency fingerprint, comprise: construction device, Real-time Obtaining device and coalignment, wherein, construction device, for extracting the advertisement audio-frequency fingerprint of advertisement audio frequency, build advertisement audio-frequency fingerprint storehouse; Real-time Obtaining device, for Real-time Obtaining live video stream, and extracts and described corresponding video flowing audio-frequency fingerprint; Coalignment, for at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint, and whether more described advertisement audio-frequency fingerprint and described video flowing audio-frequency fingerprint match, wherein, when described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint do not match, then continue search; When described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint match, record the number of times that matches of the advertisement audio frequency corresponding with advertisement audio-frequency fingerprint, if described in match number of times when exceeding specific threshold, then the moving advertising corresponding with described advertisement audio frequency is sent to client.
The synchronous put-on method of real-time moving advertising based on audio-frequency fingerprint of the present invention and equipment realize the synchronous input of moving advertising and television advertising, it is a kind of moving advertising input pattern of innovation, compare common moving advertising put-on method, the input of advertisement has more promptness (synchronous with television advertising), for advertiser, improve the repetition rate of advertisement, thus improve the probability that user buys commodity, improve the input-output ratio of advertiser; For user, meet the demand that it is checked in time and buys commodity.
Accompanying drawing explanation
By reading hereafter detailed description of the preferred embodiment, various other advantage and benefit will become cheer and bright for those of ordinary skill in the art.Accompanying drawing only for illustrating the object of preferred implementation, and does not think limitation of the present invention.And in whole accompanying drawing, represent identical parts by identical reference symbol.Wherein in the accompanying drawings, the multiple identical parts of alphabetic flag instruction after reference number, when making a general reference these parts, by its last alphabetic flag of omission.In the accompanying drawings:
Fig. 1 is that conventional ads throws in mode schematic diagram;
Fig. 2 is the process flow diagram of the synchronous put-on method of real-time moving advertising based on audio-frequency fingerprint of the present invention;
Fig. 3 is the structural representation in the audio-frequency fingerprint of advertisement described in the present invention storehouse;
Fig. 4 is the structural representation of the equipment synchronously thrown in based on the real-time moving advertising of audio-frequency fingerprint of the present invention; And
Fig. 5 is the block diagram showing the computing equipment that can be used to put into practice embodiments of the present invention.
In the accompanying drawings, use same or similar label to refer to same or similar element.
Embodiment
Illustrative embodiments of the present invention is described in detail referring now to accompanying drawing.The embodiment that to should be appreciated that shown in accompanying drawing and describe is only exemplary, is intended to explain principle of the present invention and spirit, and not limits the scope of the invention.
Fig. 2 shows the process flow diagram of the synchronous put-on method of real-time moving advertising based on audio-frequency fingerprint of the present invention.Be appreciated that the process flow diagram shown in Fig. 2 is only schematic, the step wherein recorded can perform according to different order, executed in parallel, omission and/or increase other steps.As shown in Figure 2, the described synchronous put-on method of real-time moving advertising based on audio-frequency fingerprint, comprising: the advertisement audio-frequency fingerprint extracting advertisement audio frequency, builds advertisement audio-frequency fingerprint storehouse 102; Real-time Obtaining live video stream, and extract the video flowing audio-frequency fingerprint 104 corresponding with described live video stream, described live video stream is television advertising; At described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint 106; Whether more described advertisement audio-frequency fingerprint and described video flowing audio-frequency fingerprint match 108, wherein, when described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint do not match, return step 106 item and continue at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint; When described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint match, record the number of times that matches of the advertisement audio frequency corresponding with advertisement audio-frequency fingerprint, if described in match number of times when exceeding specific threshold, then the moving advertising 110 corresponding with described advertisement audio frequency is sent to client.Described client-side is as being mobile phone, iPad, intelligent watch or panel computer.
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, extract the advertisement audio-frequency fingerprint of advertisement audio frequency, structure advertisement audio-frequency fingerprint storehouse and step 102 comprise the advertisement audio frequency 1021 extracting described moving advertising; Multiple advertisement audio-frequency fingerprint 1022 is extracted from described advertisement audio frequency; Extract key value and the value value 1023 of described advertisement audio-frequency fingerprint; Set up and have the value value of identical key value and the corresponding relation of described key value, create a Hash table, described Hash table is described advertisement audio-frequency fingerprint storehouse 1024.
Step 1021 is for extracting the advertisement audio frequency of described moving advertising.First obtain the some advertisements having the synchronous advertiser throwing in moving advertising demand and provide, moving advertising comprises Audio and Video, then from described moving advertising advertisement audio extraction out, the quantity of described advertisement audio frequency has several.
Step 1022 for extract multiple advertisement audio-frequency fingerprint from described advertisement audio frequency.The quantity of described advertisement audio frequency is several, extracts the advertisement audio-frequency fingerprint in each advertisement audio frequency, and wherein each advertisement audio frequency comprises multiple advertisement audio-frequency fingerprint.
Step 1023 is for extracting key value and the value value of described advertisement audio-frequency fingerprint.Described key value and described value value two values constitute described advertisement audio-frequency fingerprint jointly.
Step 1024 has the value value of identical key value and the corresponding relation of described key value for setting up, and create a Hash table, described Hash table is described advertisement audio-frequency fingerprint storehouse (as shown in Figure 3).For the audio-frequency fingerprint that key value is identical, can pointer be used to connect.
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, from described advertisement audio frequency, extract multiple advertisement audio-frequency fingerprint 1022 comprise: the advertisement audio file described advertisement audio frequency being preserved into consolidation form; For each advertisement audio file gives unique coding, obtained the spectrogram of each advertisement audio file by mapping algorithm, wherein, the transverse axis of described spectrogram is time t, and the longitudinal axis is frequency f; By described spectrogram according to certain transverse and longitudinal axle resolution, divide several row rectangular areas, for each region, calculate the extreme point of the highest position of each region energy as each region, wherein, described extreme point represents with transverse and longitudinal coordinate (t, f); Described advertisement audio-frequency fingerprint is obtained according to the relation between described extreme point (t, f).
Carry out description of step 1022 below by specific embodiment and namely from described advertisement audio frequency, extract multiple advertisement audio-frequency fingerprint.First use the audio frequency and video handling implement software of standard as ffmpeg, by the advertisement audio extraction in magnanimity advertisement out, preserve into the file of consolidation form, as mp3 form, wma form, wav form or ogg form.Then give unique coding for each advertisement audio file, obtained the spectrogram of each advertisement audio file by mapping algorithm, wherein, the transverse axis of described spectrogram is time t, and the longitudinal axis is frequency f, and each point value represents energy e.Then to the spectrogram of each advertisement audio file according to certain transverse and longitudinal axle resolution, divide several row rectangular areas, for each region, calculate the extreme point of the highest position of each region energy as each region, wherein, described extreme point represents with transverse and longitudinal coordinate (t, f).Finally obtain described advertisement audio-frequency fingerprint according to the relation between described extreme point (t, f).
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, obtain described advertisement audio-frequency fingerprint according to the relation between described extreme point (t, f) and comprise: obtain extreme point (t
m, f
m) and (t
n, f
n), then dT=t
n-t
m; F
n: f
m: dT as the key value of described advertisement audio-frequency fingerprint, t
mwith the value value of coding as described advertisement audio-frequency fingerprint.Such as, will be designated as the advertisement audio file of No. from coding and extract advertisement audio-frequency fingerprint, then the extreme point first obtained is (t
1, f
1) and (t
2, f
2), learn dT=t
2-t
1, then the key value of described advertisement audio-frequency fingerprint is f
2: f
1: dT, the value value of described advertisement audio-frequency fingerprint is t
1with No. one.
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, described mapping algorithm is Fourier Transform Algorithm.The spectrogram of each advertisement audio file is obtained by described Fourier Transform Algorithm.
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, Real-time Obtaining live video stream, and the extraction video flowing audio-frequency fingerprint corresponding with described live video stream and step 104 comprise: monitor live video stream, extract the video flowing audio frequency 1041 corresponding with described live video stream; Video flowing audio-frequency fingerprint 1042 is obtained according to certain time interval according to described video flowing audio frequency; Extract k value and the v value 1042 of described video flowing audio-frequency fingerprint.
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, obtain video flowing audio-frequency fingerprint according to described video flowing audio frequency according to certain time interval and comprise: described video flowing audio frequency is preserved into identical with described advertisement audio format; Obtained the spectrogram of each video flowing audio frequency by mapping algorithm, wherein, the transverse axis of described spectrogram is time t ', and the longitudinal axis is frequency f '; By described spectrogram according to certain transverse and longitudinal axle resolution, divide several row rectangular areas, for each region, calculate the extreme point of the highest position of each region energy as each region, wherein, described extreme point represents with transverse and longitudinal coordinate (t ', f '); Described video flowing audio-frequency fingerprint is obtained according to the relation between described extreme point (t ', f ').
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, obtain described video flowing audio-frequency fingerprint according to the relation between described extreme point (t ', f ') and comprise: obtain extreme point (t '
m, f '
m) and (t '
n, f '
n), then dT '=t '
n-t '
m; F '
n: f '
m: dT ' as the k value of described video flowing audio-frequency fingerprint, t '
mas the v value of described video flowing audio-frequency fingerprint.The extreme point such as first obtained for (t '
1, f '
1) and (t '
2, f '
2), learn dT '=t '
2-t '
1, then the k value of described video flowing audio-frequency fingerprint is f '
2: f '
1: dT ', the v value of described video flowing audio-frequency fingerprint is t '
1.
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, be the advertisement audio-frequency fingerprint searching the key value identical with k value in described advertisement audio-frequency fingerprint storehouse at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint.
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, whether more described advertisement audio-frequency fingerprint matches with described video flowing audio-frequency fingerprint comprises: the v value that the k value of the value value that the key value of more described advertisement audio-frequency fingerprint is corresponding and described video flowing audio-frequency fingerprint is corresponding, wherein, when v value is identical with value value, the coding of the corresponding advertisement audio frequency of described advertisement audio-frequency fingerprint is counted; Whether the counts of calculation code exceedes specific threshold.When the counts of encoding exceedes specific threshold, then the moving advertising that described live video stream is corresponding with the advertisement audio frequency of described coding is identical.Such as specific threshold can be set to 50, when the counts being encoded to No. is more than 50, then thinks that described live video stream is identical with the described moving advertising being encoded to the advertisement audio frequency of No. corresponding.Then notice has the merchandise news that the synchronous advertiser throwing in moving advertising demand is playing, and corresponding moving advertising thrown in immediately by party in request's platform (DSP).
Of the present invention based in the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, wherein, the moving advertising corresponding with described advertisement audio frequency is sent to client and comprises: on party in request's platform, throw in the moving advertising corresponding with described advertisement audio frequency; The browse advertisements of more described moving advertising and described client, wherein, when described moving advertising is relevant to described browse advertisements, is pushed to described client described moving advertising.
When thinking that the moving advertising corresponding with advertisement audio frequency of described live video stream is identical, notice has the merchandise news that the synchronous advertiser throwing in moving advertising demand is playing, and corresponding moving advertising thrown in immediately by party in request's platform (DSP).Party in request's platform allows advertiser more effectively to buy advertisement stock, because described party in request platform summarizes the stock of various advertisement transaction platform.
The browse advertisements of more described moving advertising and described client, wherein, when described moving advertising is relevant to described browse advertisements, is pushed to described client by push technology described moving advertising.Such as, inquire the advertisement that advertisement that Client browse crosses relates to automobile, described moving advertising also relates to the advertisement of automobile, then think that described moving advertising is relevant to described browse advertisements, and push technology will be pushed to client described moving advertising.In the application program of client/server, push technology can transmit data to client and send request without the need to it.
Fig. 4 is the structural representation of the synchronous dispensing device of real-time moving advertising based on audio-frequency fingerprint of the present invention.As shown in Figure 4, the described synchronous dispensing device of real-time moving advertising based on audio-frequency fingerprint comprises: construction device 10, Real-time Obtaining device 11 and coalignment 12, wherein, construction device 10, for extracting the advertisement audio-frequency fingerprint of advertisement audio frequency, builds advertisement audio-frequency fingerprint storehouse; Real-time Obtaining device 11 for Real-time Obtaining live video stream, and extracts and described corresponding video flowing audio-frequency fingerprint; Coalignment 12 is at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint, and whether more described advertisement audio-frequency fingerprint and described video flowing audio-frequency fingerprint match, wherein, when described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint do not match, then continue search; When described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint match, record the number of times that matches of the advertisement audio frequency corresponding with advertisement audio-frequency fingerprint, if described in match number of times when exceeding specific threshold, then the moving advertising corresponding with described advertisement audio frequency is pushed to client.
Fig. 5 shows the structured flowchart of the example calculation equipment that can be used for putting into practice embodiment of the present invention.Be appreciated that client, agency and/or the server recorded in present disclosure can utilize the computing equipment shown in Fig. 5 to realize.As shown in Figure 5, computing equipment can comprise: CPU (CPU (central processing unit)) 401, RAM (random access memory) 402, ROM (ROM (read-only memory)) 403, system bus 404, hard disk controller 405, keyboard controller 406, serial interface controller 407, parallel interface controller 408, display controller 409, hard disk 410, keyboard 411, serial peripheral equipment 412, concurrent peripheral equipment 413 and display 414.In these parts, what be coupled with system bus 404 has CPU401, RAM402, ROM403, hard disk controller 405, keyboard controller 406, serial interface controller 407, parallel interface controller 408 and display controller 409.Hard disk 410 is coupled with hard disk controller 405, keyboard 411 is coupled with keyboard controller 406, serial peripheral equipment 412 is coupled with serial interface controller 407, and concurrent peripheral equipment 413 is coupled with parallel interface controller 408, and display 414 is coupled with display controller 409.
The synchronous put-on method of real-time moving advertising based on audio-frequency fingerprint of the present invention and equipment realize the synchronous input of moving advertising and television advertising, it is a kind of moving advertising input pattern of innovation, compare common moving advertising put-on method, the input of advertisement has more promptness (synchronous with television advertising), for advertiser, improve the repetition rate of advertisement, thus improve the probability that user buys commodity, improve the input-output ratio of advertiser; For user, meet the demand that it is checked in time and buys commodity.
Some embodiments of the present invention are described by way of example above.Above-mentioned different masses, can being performed at least partially of operation and technology, by using hardware, processor performs firmware instructions, the instruction of processor executive software, or and combination in any.When adopting the processor performing firmware and software instruction to perform, software or firmware instructions can be stored in the storage of any computer-readable, such as disk, CD or other storage mediums, at RAM or ROM or flash storer, processor, hard disk, CD, disk etc.Similarly, software and firmware instructions can be transferred to user or system, are comprised by transmission mode that is known or that expect arbitrarily, such as, at computer readable diskette or other portable computer memory mechanisms or pass through telecommunication media.Telecommunication media typically specifically talks about computer-readable instruction, data structure, program module or other data in modulated message signal such as carrier wave or other transmission mechanisms.By example, and unrestricted, and communication media comprises wire medium such as cable network or single line and connects, and wireless medium, such as sound, wireless frequency, infrared and other wireless medium.Thus, software and firmware instructions can be transferred to user or system, by communication channel, such as telephone wire, DSL line, cable television line, optical fiber cable, wireless channel, the Internet, etc. (provide such software by portable storage media, it is counted as identical or interchangeable).Software or firmware instructions can comprise and instructions, and it causes processor to perform different action when being executed by a processor.
It should be noted that embodiments of the present invention can be realized by the combination of software, hardware or software and hardware.Hardware components can utilize special logic to realize; Software section can store in memory, and by suitable instruction execution system, such as microprocessor or special designs hardware perform.Those having ordinary skill in the art will appreciate that above-mentioned method and system and computer executable instructions can be used and/or be included in processor control routine and realize, such as, on the programmable memory of mounting medium, such as ROM (read-only memory) (firmware) or the data carrier of such as optics or electrical signal carrier of such as disk, CD or DVD-ROM, provide such code.The equipment of present embodiment and module thereof can be realized by the hardware circuit of the programmable hardware device of the semiconductor of such as VLSI (very large scale integrated circuit) or gate array, such as logic chip, transistor etc. or such as field programmable gate array, programmable logic device etc., also with the software simulating performed by various types of processor, also can be realized by the combination such as firmware of above-mentioned hardware circuit and software.
The communication network mentioned in instructions can comprise disparate networks, include but not limited to LAN (Local Area Network) (" LAN "), wide area network (" WAN "), based on IP agreement network (such as, the Internet) and ad-hoc network (such as, adhoc peer-to-peer network).
In addition, although describe the operation of the inventive method in the accompanying drawings with particular order, this is not that requirement or hint must perform these operations according to this particular order, or must perform the result that all shown operation could realize expectation.On the contrary, the step described in process flow diagram can change execution sequence.Additionally or alternatively, some step can be omitted, multiple step is merged into a step, or a step is decomposed into multiple step.
Although describe the present invention with reference to some embodiments, should be appreciated that, the present invention is not limited to disclosed embodiment.The present invention is intended to contain the interior included various amendment of spirit and scope and the equivalent arrangements of claims.The scope of claims meets the most wide in range explanation, thus comprises all such amendments and equivalent structure and function.
Claims (10)
1., based on the synchronous put-on method of real-time moving advertising of audio-frequency fingerprint, comprising:
Extract the advertisement audio-frequency fingerprint of advertisement audio frequency, build advertisement audio-frequency fingerprint storehouse;
Real-time Obtaining live video stream, and extract the video flowing audio-frequency fingerprint corresponding with described live video stream;
At described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint;
Whether more described advertisement audio-frequency fingerprint and described video flowing audio-frequency fingerprint match, wherein,
When described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint do not match, then continue at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint;
When described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint match, record the number of times that matches of the advertisement audio frequency corresponding with advertisement audio-frequency fingerprint, if described in match number of times when exceeding specific threshold, then the moving advertising corresponding with described advertisement audio frequency is pushed to client.
2. method according to claim 1, wherein, extracts the advertisement audio-frequency fingerprint of advertisement audio frequency, builds advertisement audio-frequency fingerprint storehouse and comprises:
Extract the advertisement audio frequency of described moving advertising;
Multiple advertisement audio-frequency fingerprint is extracted from described advertisement audio frequency;
Extract key value and the value value of described advertisement audio-frequency fingerprint;
Set up and have the value value of identical key value and the corresponding relation of described key value, create a Hash table, described Hash table is described advertisement audio-frequency fingerprint storehouse;
Preferably, the moving advertising corresponding with described advertisement audio frequency is sent to client to comprise:
The moving advertising corresponding with described advertisement audio frequency thrown in by party in request's platform;
The browse advertisements of more described moving advertising and described client, wherein, when described moving advertising is relevant to described browse advertisements, is pushed to described client described moving advertising.
3. method according to claim 2, wherein, from described advertisement audio frequency, extract multiple advertisement audio-frequency fingerprint comprise:
Described advertisement audio frequency is preserved into the advertisement audio file of consolidation form;
For each advertisement audio file gives unique coding, obtained the spectrogram of each advertisement audio file by Fourier Transform Algorithm, wherein, the transverse axis of described spectrogram is time t, and the longitudinal axis is frequency f;
By described spectrogram according to certain transverse and longitudinal axle resolution, divide several row rectangular areas, for each region, calculate the extreme point of the highest position of each region energy as each region, wherein, described extreme point represents with transverse and longitudinal coordinate (t, f);
Described advertisement audio-frequency fingerprint is obtained according to the relation between described extreme point (t, f).
4. method according to claim 3, wherein, obtains described advertisement audio-frequency fingerprint according to the relation between described extreme point (t, f) and comprises:
Obtain extreme point (t
m, f
m) and (t
n, f
n), then dT=t
n-t
m;
F
n: f
m: dT as the key value of described advertisement audio-frequency fingerprint, t
mwith the value value of coding as described advertisement audio-frequency fingerprint.
5. method according to claim 1, wherein, Real-time Obtaining live video stream, and the extraction video flowing audio-frequency fingerprint corresponding with described live video stream comprises:
Monitoring live video stream, extracts the video flowing audio frequency corresponding with described live video stream;
Video flowing audio-frequency fingerprint is obtained according to certain time interval according to described video flowing audio frequency;
Extract k value and the v value of described video flowing audio-frequency fingerprint.
6. method according to claim 5, wherein, obtains video flowing audio-frequency fingerprint according to described video flowing audio frequency according to certain time interval and comprises:
Described video flowing audio frequency is preserved into identical with described advertisement audio format;
Obtained the spectrogram of each video flowing audio frequency by Fourier Transform Algorithm, wherein, the transverse axis of described spectrogram is time t ', and the longitudinal axis is frequency f ';
By described spectrogram according to certain transverse and longitudinal axle resolution, divide several row rectangular areas, for each region, calculate the extreme point of the highest position of each region energy as each region, wherein, described extreme point represents with transverse and longitudinal coordinate (t ', f ');
Described video flowing audio-frequency fingerprint is obtained according to the relation between described extreme point (t ', f ').
7. method according to claim 6, wherein, obtains described video flowing audio-frequency fingerprint according to the relation between described extreme point (t ', f ') and comprises:
Acquisition extreme point (t '
m, f '
m) and (t '
n, f '
n), then dT '=t '
n-t '
m;
F '
n: f '
m: dT ' as the k value of described video flowing audio-frequency fingerprint, t '
mas the v value of described video flowing audio-frequency fingerprint.
8. method according to claim 7 wherein, is the advertisement audio-frequency fingerprint searching the key value identical with k value in described advertisement audio-frequency fingerprint storehouse at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint.
9. method according to claim 8, wherein, whether more described advertisement audio-frequency fingerprint and described video flowing audio-frequency fingerprint match and comprise:
The v value that the corresponding value value of the key value of more described advertisement audio-frequency fingerprint is corresponding with the k value of described video flowing audio-frequency fingerprint, wherein, when v value is identical with value value, counts the coding of the corresponding advertisement audio frequency of described advertisement audio-frequency fingerprint;
Whether the counts of calculation code exceedes specific threshold.
10. based on the equipment that the real-time moving advertising of audio-frequency fingerprint is synchronously thrown in, comprising: construction device, Real-time Obtaining device and coalignment, wherein,
Construction device, for extracting the advertisement audio-frequency fingerprint of advertisement audio frequency, builds advertisement audio-frequency fingerprint storehouse;
Real-time Obtaining device, for Real-time Obtaining live video stream, and extracts and described corresponding video flowing audio-frequency fingerprint;
Coalignment, at described advertisement audio-frequency fingerprint library searching advertisement audio-frequency fingerprint, and whether more described advertisement audio-frequency fingerprint and described video flowing audio-frequency fingerprint match, wherein,
When described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint do not match, then continue search;
When described video flowing audio-frequency fingerprint and described advertisement audio-frequency fingerprint match, record the number of times that matches of the advertisement audio frequency corresponding with advertisement audio-frequency fingerprint, if described in match number of times when exceeding specific threshold, then the moving advertising corresponding with described advertisement audio frequency is pushed to client.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510557641.1A CN105184610A (en) | 2015-09-02 | 2015-09-02 | Real-time mobile advertisement synchronous putting method and device based on audio fingerprints |
| KR1020160014655A KR20170027648A (en) | 2015-09-02 | 2016-02-05 | Method and apparatus for synchronous putting of real-time mobile advertisement based on audio fingerprint |
| KR1020160017280A KR20170027649A (en) | 2015-09-02 | 2016-02-15 | Method and apparatus for synchronous putting of real-time mobile advertisement based on audio fingerprint |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510557641.1A CN105184610A (en) | 2015-09-02 | 2015-09-02 | Real-time mobile advertisement synchronous putting method and device based on audio fingerprints |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105184610A true CN105184610A (en) | 2015-12-23 |
Family
ID=54906668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510557641.1A Pending CN105184610A (en) | 2015-09-02 | 2015-09-02 | Real-time mobile advertisement synchronous putting method and device based on audio fingerprints |
Country Status (2)
| Country | Link |
|---|---|
| KR (2) | KR20170027648A (en) |
| CN (1) | CN105184610A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105574193A (en) * | 2015-12-25 | 2016-05-11 | 北京奇虎科技有限公司 | Method, device and system for acquiring related information of advertisement |
| CN105701686A (en) * | 2016-01-23 | 2016-06-22 | 北京掌阔移动传媒科技有限公司 | Voiceprint advertisement implementation method and device |
| CN105933761A (en) * | 2016-06-24 | 2016-09-07 | 中译语通科技（北京）有限公司 | Novel audio-visual program advertisement broadcasting method |
| CN106055570A (en) * | 2016-05-19 | 2016-10-26 | 中国农业大学 | Video retrieval device based on audio data and video retrieval method for same |
| CN106101573A (en) * | 2016-06-24 | 2016-11-09 | 中译语通科技（北京）有限公司 | The grappling of a kind of video labeling and matching process |
| CN106375867A (en) * | 2016-08-22 | 2017-02-01 | 无锡天脉聚源传媒科技有限公司 | Method and device for cutting advertisement in television video |
| CN107240160A (en) * | 2016-03-28 | 2017-10-10 | 富士施乐株式会社 | Information processing system, information processing method and electronic installation |
| CN107346665A (en) * | 2017-06-29 | 2017-11-14 | 广州视源电子科技股份有限公司 | Method, apparatus, equipment and the storage medium of audio detection |
| CN107633848A (en) * | 2016-07-19 | 2018-01-26 | 阿基米德（上海）传媒有限公司 | It is a kind of to remove the method for advertisement by identifying audio-frequency fingerprint |
| CN108920529A (en) * | 2018-06-07 | 2018-11-30 | 北京金山安全软件有限公司 | Data extraction method and device and electronic equipment |
| CN109977630A (en) * | 2017-12-27 | 2019-07-05 | 上海全土豆文化传播有限公司 | Audio-frequency fingerprint generation method and device |
| CN113129855A (en) * | 2019-12-30 | 2021-07-16 | 阿里巴巴集团控股有限公司 | Audio fingerprint extraction and database building method, and audio identification and retrieval method and system |
| CN114866827A (en) * | 2022-01-07 | 2022-08-05 | 广州繁星互娱信息科技有限公司 | Audio and video synchronization detection method and device, storage medium and electronic equipment |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108198573B (en) * | 2017-12-29 | 2021-04-30 | 北京奇艺世纪科技有限公司 | Audio recognition method and device, storage medium and electronic equipment |
| US20210157838A1 (en) * | 2019-11-26 | 2021-05-27 | Gracenote, Inc. | Methods and apparatus to fingerprint an audio signal via exponential normalization |
| US11798577B2 (en) | 2021-03-04 | 2023-10-24 | Gracenote, Inc. | Methods and apparatus to fingerprint an audio signal |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799605A (en) * | 2012-05-02 | 2012-11-28 | 天脉聚源(北京)传媒科技有限公司 | Method and system for monitoring advertisement broadcast |
| CN102831537A (en) * | 2012-07-09 | 2012-12-19 | 北京十分科技有限公司 | Method and device for obtaining network advertisement information |
| CN103794209A (en) * | 2014-01-17 | 2014-05-14 | 王博龙 | System for monitoring and playing advertisement before movie playing based on audio fingerprint identification technology |
| US20140277641A1 (en) * | 2013-03-15 | 2014-09-18 | Facebook, Inc. | Managing Silence In Audio Signal Identification |
| CN104317967A (en) * | 2014-11-17 | 2015-01-28 | 北京航空航天大学 | Two-layer advertisement audio retrieval method based on audio fingerprints |
| CN104636474A (en) * | 2015-02-13 | 2015-05-20 | 王磊 | Method and equipment for establishment of audio fingerprint database and method and equipment for retrieval of audio fingerprints |
-
2015
- 2015-09-02 CN CN201510557641.1A patent/CN105184610A/en active Pending
-
2016
- 2016-02-05 KR KR1020160014655A patent/KR20170027648A/en unknown
- 2016-02-15 KR KR1020160017280A patent/KR20170027649A/en not_active Application Discontinuation
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799605A (en) * | 2012-05-02 | 2012-11-28 | 天脉聚源(北京)传媒科技有限公司 | Method and system for monitoring advertisement broadcast |
| CN102831537A (en) * | 2012-07-09 | 2012-12-19 | 北京十分科技有限公司 | Method and device for obtaining network advertisement information |
| US20140277641A1 (en) * | 2013-03-15 | 2014-09-18 | Facebook, Inc. | Managing Silence In Audio Signal Identification |
| CN103794209A (en) * | 2014-01-17 | 2014-05-14 | 王博龙 | System for monitoring and playing advertisement before movie playing based on audio fingerprint identification technology |
| CN104317967A (en) * | 2014-11-17 | 2015-01-28 | 北京航空航天大学 | Two-layer advertisement audio retrieval method based on audio fingerprints |
| CN104636474A (en) * | 2015-02-13 | 2015-05-20 | 王磊 | Method and equipment for establishment of audio fingerprint database and method and equipment for retrieval of audio fingerprints |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105574193A (en) * | 2015-12-25 | 2016-05-11 | 北京奇虎科技有限公司 | Method, device and system for acquiring related information of advertisement |
| CN105701686A (en) * | 2016-01-23 | 2016-06-22 | 北京掌阔移动传媒科技有限公司 | Voiceprint advertisement implementation method and device |
| CN107240160A (en) * | 2016-03-28 | 2017-10-10 | 富士施乐株式会社 | Information processing system, information processing method and electronic installation |
| CN106055570A (en) * | 2016-05-19 | 2016-10-26 | 中国农业大学 | Video retrieval device based on audio data and video retrieval method for same |
| CN105933761A (en) * | 2016-06-24 | 2016-09-07 | 中译语通科技（北京）有限公司 | Novel audio-visual program advertisement broadcasting method |
| CN106101573A (en) * | 2016-06-24 | 2016-11-09 | 中译语通科技（北京）有限公司 | The grappling of a kind of video labeling and matching process |
| CN105933761B (en) * | 2016-06-24 | 2019-02-26 | 中译语通科技股份有限公司 | A kind of novel audio-visual program commercial throwing broadcasting method |
| CN107633848A (en) * | 2016-07-19 | 2018-01-26 | 阿基米德（上海）传媒有限公司 | It is a kind of to remove the method for advertisement by identifying audio-frequency fingerprint |
| CN106375867A (en) * | 2016-08-22 | 2017-02-01 | 无锡天脉聚源传媒科技有限公司 | Method and device for cutting advertisement in television video |
| CN107346665A (en) * | 2017-06-29 | 2017-11-14 | 广州视源电子科技股份有限公司 | Method, apparatus, equipment and the storage medium of audio detection |
| CN109977630A (en) * | 2017-12-27 | 2019-07-05 | 上海全土豆文化传播有限公司 | Audio-frequency fingerprint generation method and device |
| CN109977630B (en) * | 2017-12-27 | 2021-12-24 | 阿里巴巴（中国）有限公司 | Audio fingerprint generation method and device |
| CN108920529A (en) * | 2018-06-07 | 2018-11-30 | 北京金山安全软件有限公司 | Data extraction method and device and electronic equipment |
| CN113129855A (en) * | 2019-12-30 | 2021-07-16 | 阿里巴巴集团控股有限公司 | Audio fingerprint extraction and database building method, and audio identification and retrieval method and system |
| CN114866827A (en) * | 2022-01-07 | 2022-08-05 | 广州繁星互娱信息科技有限公司 | Audio and video synchronization detection method and device, storage medium and electronic equipment |
| CN114866827B (en) * | 2022-01-07 | 2024-04-16 | 广州繁星互娱信息科技有限公司 | Audio and video synchronization detection method and device, storage medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| KR20170027648A (en) | 2017-03-10 |
| KR20170027649A (en) | 2017-03-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105184610A (en) | Real-time mobile advertisement synchronous putting method and device based on audio fingerprints | |
| EP3779841B1 (en) | Method, apparatus and system for sending information, and computer-readable storage medium | |
| US9367603B2 (en) | Systems and methods for behavioral segmentation of users in a social data network | |
| US20160180402A1 (en) | Method for recommending products based on a user profile derived from metadata of multimedia content | |
| CN105447730A (en) | Target user orientation method and device | |
| CN111882361A (en) | Audience accurate advertisement pushing method and system based on artificial intelligence and readable storage medium | |
| WO2014124310A1 (en) | A system for improving shape-based targeting by using interest level data | |
| CN104751354B (en) | A kind of advertisement crowd screening technique | |
| CN102073956A (en) | Data mining-based directional advertisement release method, system and equipment | |
| US11288240B1 (en) | Data learning and analytics apparatuses, methods and systems | |
| CN108805598A (en) | Similarity information determines method, server and computer readable storage medium | |
| WO2017087833A1 (en) | Measuring influence propagation within networks | |
| CN105760526A (en) | News classification method and device | |
| CN106611353B (en) | Method for acquiring audience and server equipment | |
| CN107220745B (en) | Method, system and equipment for identifying intention behavior data | |
| CN112287238B (en) | User characteristic determination method and device, storage medium and electronic equipment | |
| CN107977678A (en) | Method and apparatus for output information | |
| CN112258254A (en) | Internet advertisement risk monitoring method and system based on big data architecture | |
| Lipyanina et al. | Targeting Model of HEI Video Marketing based on Classification Tree. | |
| CN111787042B (en) | Method and device for pushing information | |
| Sangaralingam et al. | High Value Customer Acquisition & Retention Modelling–A Scalable Data Mashup Approach | |
| CN111581429A (en) | Music pushing method, device, equipment and computer readable storage medium | |
| CN113672816B (en) | Account feature information generation method and device, storage medium and electronic equipment | |
| Park et al. | Multimodal learning model based on video–audio–chat feature fusion for detecting e-sports highlights | |
| CN110287315A (en) | Public sentiment determines method, apparatus, equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20151223 |
|
| RJ01 | Rejection of invention patent application after publication |