KR20100064423A - Method and system for advertising according to audio of moving picture on line - Google Patents

Abstract

PURPOSE: An online advertisement providing method according to the video audio information and a system thereof are provided to provide an advertisement matched with the audio information on a real time basis by analyzing the extracted audio information of video. CONSTITUTION: A web server(12) provides a web page to a user terminal connected through network. The web server provides the basic menu for the execution of the video contents file. A voice processing server(14) compares the voice signal transmitted from the user terminal with the advance saved voice information. The voice processing server distinguishes the actor equivalent under the voice. An advertisement server(16) provides the advertisement contents file matched with the distinguished actor.

Description

Method and system for providing online advertisement based on video audio information {Method and system for advertising according to audio of moving picture on line}

The present invention relates to a method and a system for providing an online advertisement, and more particularly, to a method and system for providing an online advertisement that can provide an advertisement matching the same in real time by extracting and analyzing audio (voice) information of a video. will be.

With the spread and development of the Internet, the online advertising market is growing. Recently, according to the trend of flooding of UCC and various kinds of multimedia contents based on Web 2.0, online advertisement technologies using them have been proposed.

Korean Patent Publication No. 2008-88437 shows a video advertisement insertion system that caches original multimedia content in real time and inserts a high quality video advertisement directly from a cache server.

However, such a system merely provides by pushing a predetermined advertisement irrespective of the contents of the multimedia content.

For example, when playing multimedia content such as a specific drama or movie, a so-called target advertisement that provides advertisements associated with the actor or situation in the drama or movie may be very effective.

According to the related art, in order to implement the target advertisement as described above, it is necessary to pre-analyze the actors in the individual video contents, the situation or the dialogue, and synchronize the advertisements according to the timing of the frame in which the scene appears. .

Therefore, there is a limitation in providing various advertisements in accordance with a large amount of multimedia content because it is necessary to perform a very cumbersome and time-consuming predecessor for the target advertisement.

The present invention is to solve the above problems, to provide a method and system for providing an online advertisement that can extract the audio signal from the audio signal included in the video content file and display an advertisement matching the specific actor or sound effect therefrom. It aims to do it.

In order to achieve the above object, an online advertisement providing method according to a preferred embodiment of the present invention comprises the steps of: (a) receiving a video content file including a video signal and an audio signal from a server and playing it on a user terminal; (b) extracting and transmitting an audio / audio signal of the played video content file to the server; (c) performing a Fourier transform on the extracted audio and audio signals in the server and dividing the extracted audio and audio signals into a plurality of sub-signals having different frequencies; (d) comparing the sub-signals with pre-stored voice information to determine an actor corresponding to the voice; (e) extracting an advertisement content file matching the actor from the server and transmitting the extracted advertisement content file to the user terminal; And (f) executing and displaying the transmitted advertisement content file in the user terminal.

Preferably, in the step (d), the frequency is compared in order from the sub-signal having the maximum amplitude to the sub-signal with the minimum amplitude.

More preferably, the frequency of the sub-signals is obtained by grasping the period of the intersection where each sub-signal meets the yin-yang reference line.

According to another embodiment of the present invention, in an online advertisement providing method using a system in which a user terminal and a server are connected through a network, (a) receiving a video content file including a video signal and an audio signal from the server; Playing on the user terminal; (b) extracting and transmitting the sound effect of the played video content file to the server; (c) performing a Fourier transform on the extracted sound effect at the server and dividing the extracted sound into a plurality of sub-signals having different frequencies; (d) comparing the sub-signals with previously stored sound effect information to determine the type of the sound effect; (e) extracting an advertisement content file matching the type of the sound effect from the server and transmitting the extracted advertisement content file to the user terminal; And (f) executing and displaying the transmitted advertisement content file in the user terminal.

Furthermore, the online advertisement providing method of the present invention comprises the steps of: converting a voice signal transmitted from the user terminal to a server into text; Searching for the converted text to determine whether there is text that matches a pre-stored advertisement keyword; If there is text that matches the advertisement keyword, extracting the corresponding advertisement content file and transmitting it to the user terminal; And executing and displaying the transmitted advertisement content file in the user terminal.

According to another aspect of the present invention, a web server for providing a web page to a user terminal connected through a network and providing a basic menu for executing a video content file; A speech processing server that determines an actor corresponding to the speech by comparing the speech signal transmitted from the user terminal with previously stored speech information; And an advertisement server extracting an advertisement content file matching the determined actor and providing the same to the user terminal.

Preferably, the voice processing server comprises: a Fourier transform unit for dividing a voice signal transmitted from the user terminal into a plurality of sub-signals having different frequencies; And a signal comparator for comparing the divided sub-signals with pre-stored voice information and frequencies in order from sub-signals having a maximum amplitude to sub-signals with a minimum amplitude.

More preferably, the voice processing server, STT conversion unit for converting the voice signal transmitted from the user terminal to the text; And a keyword search unit for searching the converted text to determine whether there is text that matches the pre-stored advertisement keyword.

According to the online advertisement providing method of the present invention, since the voice of the actor appearing in the video being played on the user terminal is provided in real time and the advertisement is matched with it, the pre-work of matching the advertisement by analyzing the video frame one by one as before. This is unnecessary.

That is, since the advertising content can be independently produced and provided regardless of the video content, it is possible to secure a wide range of advertisement targets and ranges, and also has an excellent advertising effect.

In addition, according to the online advertisement providing method of the present invention, not only the actors appearing in the video, but also the advertisements matching the same may be provided for the sound effects such as the sound of water, the sound of a car, the sound of an airplane, and the like.

Furthermore, the present invention can achieve a further enhanced effect in parallel with the embodiment of converting the voice into text and extracting the advertising keyword accordingly to provide a matching advertisement.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Prior to this, terms or words used in the specification and claims should not be construed as having a conventional or dictionary meaning, and the inventors should properly explain the concept of terms in order to best explain their own invention. Based on the principle that can be defined, it should be interpreted as meaning and concept corresponding to the technical idea of the present invention. Therefore, the embodiments described in the specification and the drawings shown in the drawings are only the most preferred embodiment of the present invention and do not represent all of the technical idea of the present invention, various modifications that can be replaced at the time of the present application It should be understood that there may be equivalents and variations.

System overview

Figure 1 shows a schematic configuration of a system for implementing a method for providing online advertising according to a preferred embodiment of the present invention.

Referring to the drawings, a system according to the present invention processes a voice signal extracted from a video content file and provides an advertisement accordingly, and accesses the server 10 through a network 20 to the video content. It includes a user terminal 30 that executes the file and simultaneously displays an advertisement.

The server 10 includes a web server 12 that provides a basic web page to a user terminal 30 connected through a network 20 and provides a basic menu for executing a member login or a video content file. And a voice processing server 14 for processing a voice signal extracted from the video content file, and an advertisement server 16 for providing a predetermined advertisement content file to the user terminal 30 according to the result of the voice processing.

The web server 12 provides, for example, a web page written in HTML to the user terminal 30 connected through the Internet, and the web page is provided with a web browser 32 (eg, provided in the user terminal 30). Is interpreted and displayed by Internet Explorer.

In addition, the web page provided by the web server 12 provides a user interface (not shown) for member registration and login, and the user may register or perform a login procedure through the web page. .

The voice processing server 14 processes a voice signal transmitted from the user terminal 30 and then compares the voice signal with the voice information data stored in the voice information DB 42 to identify a specific actor or a specific sound effect. Let's do it.

The advertisement server 16 transmits the advertisement content file corresponding to the specific actor or the specific sound effect to the user terminal 30 according to the result processed by the speech processing server 14, and the advertisement content file is the user terminal 30. Is displayed in a predetermined position by the advertisement display means 38 provided in the.

According to the present invention, a video DB 40, a voice information DB 42, and an advertisement DB 44 interworking with the server 10 are included. The video DB 40 includes various types of information to be provided to the user terminal 30. The video content file is stored.

The video content file according to the present invention includes various types of media video files such as movies, dramas, UCCs, and news including video information and audio information compressed according to a standard such as MPEG.

In the voice information DB 42, sampling data of voice waveforms and frequencies of specific actors are stored. Sampling data for such speech will be described in detail later.

The advertisement DB 44 stores various advertisement content files that are transmitted to and displayed on the user terminal 30. According to the present invention, the advertisement content file is classified and stored according to actors or sound effects. For example, as shown in FIG. 3, each advertisement information includes the name of an actor or person appearing in the advertisement, a keyword matching the advertisement, and an advertisement content file.

The advertisement content file includes not only a flash or a video but also an image file or a text file such as GIF or JPEG.

Therefore, as will be described later, the advertisement content file is provided when the actor's voice matching it is played in the video.

The user terminal 30 includes a personal terminal capable of internet access, such as a conventional personal PC, PDA terminal, PMP terminal, DMB terminal and Wibro terminal. In the user terminal 30, for example, a web browser 32 is installed that can access a web server 12 and interpret and display web pages such as HTML, XML, HDML, etc. provided by the server through a wired or wireless Internet. And an independent client program is installed and executed.

In addition, the user terminal 30 is further provided with a video player 34, an audio extraction means 36 and an advertisement display means 38.

The video player 34 executes a video content file provided from the server 10. For example, a Windows Media Player of Microsoft Corporation and a Gomplayer of Gretech. Or adrenalin or a separate client program provided by the server 10.

The audio extraction means 36 extracts only an audio (audio) signal from a video signal, an audio signal, head information, etc. of a video content file provided from the server 10 to the user terminal 30 and transmits the audio signal to the server 10.

The advertisement display means 38 performs a function of displaying the advertisement content file provided by the advertisement server 16 in a predetermined area of the user terminal 30.

Each component of the server and the user terminal is not divided in terms of physical and hardware concepts but according to the functions they perform, and these functions may be implemented by application software programmed to perform them. Ruleon, the server and the user terminal have a basic display device such as a CRT or LCD for screen display, and an input means such as a keyboard for information input.

The network 20 connecting the server 10 and the user terminal 30 includes a mobile communication network including the Internet, a public telephone network (PSTN), a cellular phone, a PCS, and a TRS, and a wireless network connecting fixed communication means to a wireless network. It is defined as a comprehensive concept including a subscriber network (Wireless Local Loop).

In the present specification and claims, “means” should be understood as program units that perform certain functions rather than physical components.

Configuration of Speech Processing Server

As shown in FIG. 2, the voice processing server 14 according to the present invention includes a Fourier transform unit 52, a signal comparator 54, an STT converter 56, and a keyword searcher 58.

The Fourier transform unit 52 performs a function of dividing into a frequency spectrum in order to analyze the voice signal extracted by the audio extraction means 36.

According to the present invention, the advertisement can be performed in real time by extracting the voice of a specific actor. In this case, the voice signal is analyzed to distinguish the unique voice according to the actor. Then, the process of obtaining the voice information by analyzing the voice signal according to the present invention will be described in detail.

Voice information

Human voice is composed of high frequency and low frequency of various waveforms. Voices are distinguished from one person according to the characteristics of the synthesized wave, and even the vocal cord simulation cannot mimic the unique voice.

As described above, the speech signal, which is a synthesized wave, may be divided into a plurality of frequency spectrums through a Fourier transform. 4 is a graph schematically showing a plurality of signal waveforms obtained by Fourier transforming an audio signal. Referring to the figure, it can be seen that the voice signal is divided into several sub-signals having different frequencies.

The sub-signals have their respective waveforms, amplitudes and frequencies, and these characteristics determine what is called 'voice'.

According to a preferred embodiment of the present invention, it is possible to grasp the characteristics of the tone by analyzing the frequency of the low frequency sub-signal S1 having the maximum amplitude among the sub-signals. The reason why the sub-signals S1 having the maximum amplitude are compared is because these sub-signals take the greatest weight in determining the tone.

That is, referring to the graph shown in FIG. 5, representative voice information is obtained by identifying the period (frequency) of the intersection point where the sub-signal S1 having the maximum amplitude meets the yin-yang reference line zero.

Subsequently, the frequency can be grasped in the same manner for the sub-signals ranging from the second largest amplitude signal (secondary amplitude signal) to the smallest amplitude signal.

As shown in FIG. 6, such voice information is divided into specific actors and stored in the voice information DB 42.

Preferably, in addition to the frequency of the sub-signal, the amplitude information for each sub-signal may also be recorded and stored.

Referring back to FIG. 2, the signal comparator 54 compares the sub-signals extracted by the audio extraction means 36 and then divided by the Fourier transform unit 52 with previously stored voice information. Performs a function to find out if the voice of the.

The STT conversion unit 56 is a so-called 'Speech To Text' conversion unit for converting the audio signal extracted and transmitted by the audio extraction unit 36 into text.

In addition, the keyword search unit 58 performs a function of determining whether the advertisement keyword is included by searching the text converted by the STT conversion unit 56. If the text search result includes a predetermined advertising keyword, an advertisement corresponding to the keyword is provided.

Then, a method for providing online advertisements according to a preferred embodiment of the present invention using a system having the above configuration will be described in detail. 7 is a flowchart illustrating a process of an online advertisement providing method according to a preferred embodiment of the present invention.

Referring to the drawings together, first, a user for using the service according to the present invention accesses the server 10 connected to the network 20 using his user terminal 30. (Step S100)

Then, the web server 12 transmits a web page including a login and service menu to the user terminal 30, and the user terminal 30 interprets it by the web browser 32 to display an LCD monitor (not shown). Display on a display device such as

If the user is an existing member, the member ID and password are inputted and logged in to the server. If the user is a new member, the registration process can be performed according to the member registration screen provided separately.

The user requests his / her desired video content from the server 10 by clicking on the icon or button provided on the web page (step S110).

In response, the server 10 extracts the video content file requested by the user from the video DB 40 and transmits the video content file to the user terminal (step S120).

Then, the video player 34 of the user terminal 30 executes the transmitted video content file and displays it as shown in FIG. 8 (step S130).

As described above, when the video content file is played, the audio extraction means 36 of the user terminal 30 separates and extracts an audio (audio) signal from the video content file in real time and transmits the same to the server 10 (step S140).

The voice signal transmitted to the server is processed by the voice processing server 14. Specifically, the Fourier transform unit 52 of the voice processing server 14 converts the voice signal into a frequency spectrum and has a different frequency. The signal is divided into a plurality of sub-signals (step S150).

The divided sub-signals are compared with the information stored in the voice information DB 42 by the signal comparator 54 to determine whether there are actors having the same voice characteristics (step S160).

Specifically, the signal comparison unit 54 compares the sub-signal S1 having the maximum amplitude among the extracted voice signals with the voice information data to determine whether the same frequency is present.

Then, when only one voice information equal to the frequency of the sub-signal S1 having the maximum amplitude is searched, it is determined that an actor (eg, Kim Tae-hee) corresponding to the voice information is matched.

If more than one voice information equal to the frequency of the sub-signal having the maximum amplitude is searched for, the frequency is searched for the next sub-signal having the second largest amplitude.

In this process, each sub-signal is compared with the voice information in order from the sub-signal having the largest amplitude to the smallest amplitude, and the matching actor information is found therefrom.

In this case, a criterion of how many amplitudes to sequentially contrast may be appropriately set as necessary.

In the above process, when the actor information having the same voice information as the extracted voice is found, the advertisement server 16 then extracts the advertisement content file in which the actor appears from the advertisement DB 44 and transmits it to the user terminal 30. (Step S170)

Subsequently, the transmitted advertisement content file is displayed at a predetermined position by the advertisement display means 38 (step S180).

9 is a screen illustrating an example of displaying the advertisement content file. The advertisement content file may be made of text, an image, a video, a flash, or the like, and may be displayed at a suitable position of a web page or a player screen on which a video is executed under the control of the advertisement display means 38.

In this case, the advertisement display method may be formed in various ways such as being displayed as text on the video which is disappeared or played after being exposed for a predetermined time (5 to 10 seconds).

According to another preferred embodiment of the present invention, by extracting the keywords included in the dialogue spoken by the actor of the video together, it is possible to provide an advertisement specifically matching the video being played.

To this end, referring again to FIG. 7, the voice signal transmitted from the audio extraction means 36 of the user terminal 30 converts the voice into text by the STT converter 56 of the voice processing server 14. (Step S190) At this time, since a specific technique for converting a voice into text has been well known, a detailed description thereof will be omitted.

Subsequently, the information converted into text is processed by the keyword search unit 58 to determine whether there are words, phrases, clauses, etc. that match the advertisement keywords previously stored in the advertisement DB 44. (Step S200). )

As a result, if there is a text (for example, a bouquet) that matches a predetermined keyword, the advertisement server 16 extracts the advertisement content file corresponding to the predetermined advertisement keyword and provides it to the user terminal 30 (step S210), which is an advertisement display means. Is displayed by 38.

More preferably, according to the present invention, a predetermined advertisement may be provided when the voice of a specific actor appearing in the video and the text included in the dialogue spoken by the actor coincide with each other.

According to another preferred embodiment of the present invention, a matching advertisement is provided according to not only the voice signal of the actor, but also sound effects such as water sound, wind sound, car sound, and airplane sound included in the video (hereinafter referred to as 'effect sound'). can do.

Since such sound effects can also be distinguished by unique tones, the advertisement service can be implemented in the same manner.

In this case, the advertisement information has a predetermined advertisement content file corresponding to the specific sound effect, as shown in FIG.

For example, in the video, when the sound comes out of the plane taking off, the audio extraction means 36 transmits the sound effect signal to the server 10, and the voice processing server 14 performs Fourier transform on each of them as described above. Determine the frequency characteristics of

Then, by comparing the sound effect information data stored in advance by the signal comparator 54 to determine which sound is the sound effect, the advertisement server 16 extracts a matched advertisement content file (eg, travel agency advertisement). To be provided to the user terminal 30 to be displayed.

Therefore, in the present invention, it is possible to provide the target advertisement according to various sound effects included in the video.

Although the present invention will be described in detail with reference to the following drawings, these drawings illustrate preferred embodiments of the present invention, and the technical concept of the present invention is not limited to the drawings and should not be interpreted.

1 is a block diagram schematically illustrating a configuration of a system for implementing an online advertisement providing method according to a preferred embodiment of the present invention.

2 is a block diagram schematically illustrating a configuration of a voice processing server in an online advertisement providing system according to a preferred embodiment of the present invention.

3 is a diagram schematically showing an example of advertisement information data according to the online advertisement providing method according to a preferred embodiment of the present invention.

4 is a graph illustrating an example of splitting a voice signal into sub-signals by performing Fourier transform in an online advertisement providing method according to an exemplary embodiment of the present invention.

FIG. 5 is a diagram for describing a process of extracting a frequency of a signal having a maximum amplitude among the sub-signals shown in FIG. 4.

6 is a diagram schematically showing an example of voice information data in the online advertisement providing method according to a preferred embodiment of the present invention.

7 is a flowchart illustrating a method for providing online advertisements according to a preferred embodiment of the present invention.

8 is a diagram illustrating a screen example of playing a video content file according to the online advertisement providing method according to an exemplary embodiment of the present invention.

9 illustrates an example of displaying an advertisement content file according to an online advertisement providing method according to a preferred embodiment of the present invention.

10 is a diagram schematically showing an example of advertisement information data according to an online advertisement providing method according to another preferred embodiment of the present invention.

Claims (8)

In the online advertisement providing method using a system in which the user terminal and the server is connected to the network, (a) receiving a video content file including a video signal and an audio signal from the server and playing the video content file in the user terminal; (b) extracting and transmitting an audio / audio signal of the played video content file to the server; (c) performing a Fourier transform on the extracted audio and audio signals in the server and dividing the extracted audio and audio signals into a plurality of sub-signals having different frequencies; (d) comparing the sub-signals with pre-stored voice information to determine an actor corresponding to the voice; (e) extracting an advertisement content file matching the actor from the server and transmitting the extracted advertisement content file to the user terminal; And (f) executing and displaying the transmitted advertisement content file in the user terminal. The method of claim 1, In the step (d), A method for providing online advertisements, wherein the frequencies are compared in order from the sub-signal having the largest amplitude to the sub-signal having the smallest amplitude. The method of claim 2, The frequency of the sub-signals is obtained by grasping the period of the intersection point where each sub-signal meets the yin-yang reference line. In the online advertisement providing method using a system in which the user terminal and the server is connected to the network, (a) receiving a video content file including a video signal and an audio signal from the server and playing the video content file in the user terminal; (b) extracting and transmitting the sound effect of the played video content file to the server; (c) performing a Fourier transform on the extracted sound effect at the server and dividing the extracted sound into a plurality of sub-signals having different frequencies; (d) comparing the sub-signals with previously stored sound effect information to determine the type of the sound effect; (e) extracting an advertisement content file matching the type of the sound effect from the server and transmitting the extracted advertisement content file to the user terminal; And (f) executing and displaying the transmitted advertisement content file in the user terminal. The method according to any one of claims 1 to 4, Converting a voice signal transmitted from the user terminal to a server into text; Searching for the converted text to determine whether there is text that matches a pre-stored advertisement keyword; If there is text that matches the advertisement keyword, extracting the corresponding advertisement content file and transmitting it to the user terminal; And And executing and displaying the transmitted advertisement content file in the user terminal. A web server providing a web page to a user terminal connected through a network and providing a basic menu for executing a video content file; A speech processing server that determines an actor corresponding to the speech by comparing the speech signal transmitted from the user terminal with previously stored speech information; And And an advertisement server extracting an advertisement content file matching the determined actor and providing the same to the user terminal. The method of claim 6, The voice processing server, A Fourier transformer for dividing the voice signal transmitted from the user terminal into a plurality of sub-signals having different frequencies; And And a signal comparison unit comparing the divided sub-signals with pre-stored voice information and frequencies in order from sub-signals having a maximum amplitude to sub-signals with a minimum amplitude. The method of claim 7, wherein The voice processing server, An STT conversion unit converting the voice signal transmitted from the user terminal into text; And And a keyword search unit for searching the converted text to determine whether there is a text that matches a pre-stored advertisement keyword.

Images