KR102650533B1 - Video processing and audio output SVS(Singing Voice Synthesis) system based on blockchain copy protection and AI deep learning - Google Patents

Abstract

The present invention relates to an image processing voice/music output SVS system using blockchain-based copy prevention and deep learning AI. A content production server that produces and stores voice data of celebrities, and a content processing server that extracts syllable voices from the voice data and replaces the member's name in the congratulatory song or video content with syllable voices; , If there is a request from the service server, it may include an additional service server that provides a congratulatory song or video processed by the content processing server to the member terminal.

Description

Video processing and audio output SVS (Singing Voice Synthesis) system based on blockchain copy protection and AI deep learning}

The present invention relates to an image processing voice/music output SVS system using blockchain-based copy prevention and deep learning AI, and more specifically, to a system that can provide various services using videos and voices of celebrities.

Recently, as the entertainment business develops, the general public has a desire to communicate directly with their favorite star, and to satisfy this desire, the number of people participating in fan signing events, ARS voice mail service, etc. is increasing. .

In this way, unlike existing models, such as the production of product advertisements using celebrities as models, systems and business models that provide various services by connecting ordinary individuals with celebrities have been proposed.

For example, Registered Patent No. 10-2418180 (registered on July 4, 2022, Anniversary Celebration Service Provision System) describes a technology that allows the general public to receive a congratulatory message created directly from a specific celebrity on their anniversary.

However, the above-mentioned registered patent had the inconvenience of having to apply for the service by selecting the celebrity who would like to be congratulated, selecting the content to be congratulated, the congratulation time, etc., on the beneficiary customer terminal.

In addition, the celebrity selected on the beneficiary customer terminal is configured to check the schedule to determine whether video production is possible and approve the provision of congratulatory service only if possible, resulting in cases where service applications using the beneficiary customer terminal are actually cancelled.

When service applications are canceled like this, customers have problems such as lowering their trust in the service.

The problem that the present invention aims to solve in consideration of the above problems is to provide an image processing, voice/music output SVS system using blockchain-based copy prevention and deep learning AI that can provide services regardless of the celebrity's schedule. .

Specifically, the present invention obtains celebrity information preferred by members signing up for the service, learns the celebrity's voice, reproduces each member's name with the celebrity's voice, and inserts the celebrity's voice into a congratulatory song. , The goal is to provide a system that can provide services regardless of the celebrity's schedule.

In addition, the present invention provides a system that can improve service utilization by combining with various services.

In addition, we are improving the delivery method of congratulatory messages and providing a system that can increase member satisfaction by providing content that cannot be copied or edited.

The present invention system for solving the technical problems described above includes a service server that provides main services to members registered through member terminals, produces and stores the produced congratulatory song or video content, and produces voice data of celebrities. and a content production server that stores the content, a content processing server that extracts a syllable voice from the voice data and replaces the member's name part of the congratulatory song or video content with a syllable voice, and, if there is a request from the service server, the content. It may include an additional service server that provides a congratulatory song or video processed in a processing server to the member terminal.

In an embodiment of the present invention, the celebrity's voice data may be obtained by storing the voice of the celebrity repeatedly reading a script containing a number of sentences as digital data.

In an embodiment of the present invention, the content processing server divides the voice data into syllables, classifies the voice of each syllable through artificial intelligence learning, and then determines a characteristic representative syllable voice.

In an embodiment of the present invention, the congratulatory song or video content is a song of a celebrity or an ordinary person stored as digital data, and the part containing the member's name in the congratulatory song may be distinguished by repetition of promised syllables.

In an embodiment of the present invention, the additional service server may process congratulatory song or video content into a non-fungible token and provide it to the member terminal.

In an embodiment of the present invention, the additional service server provides congratulatory songs or video content to member terminals using app push, direct message, or messaging, or provides downloadable content from the service page provided by the additional service server. You can.

In an embodiment of the present invention, the service server may be an online shopping mall server, an exchange voucher shopping mall server, a credit card company server, or a bank server.

The present invention's blockchain-based copy prevention and deep learning AI image processing voice/music output SVS system utilizes deep learning AI to obtain celebrity information preferred by members subscribing to the service, learn the voice of the celebrity, and convert the individual member's name into the celebrity's name. It is reproduced by voice, and the celebrity's voice is inserted into the congratulatory song to provide members with a congratulatory voice message regardless of the celebrity's schedule, which has the effect of increasing the satisfaction of members (consumers).

In addition, the present invention provides an additional service that is easy to combine with various known services, which has the effect of improving the service utilization rate and helping to increase the service provider's profits.

More specifically, the present invention is applied to the app card service and provides a file containing the voice of a celebrity preferred by the app card user according to the app card usage performance, which has the effect of increasing the app card usage performance and increasing profits.

The present invention provides congratulatory messages to members in the form of direct messages, chatting applications, etc., but provides them in the form of NFTs, thereby strengthening security and making duplication and editing impossible, which has the effect of increasing member satisfaction. there is.

Figure 1 is a block diagram of an image processing voice/music output SVS system using blockchain-based copy prevention and deep learning AI according to a preferred embodiment of the present invention.
FIG. 2 is a block diagram of the content processing server in FIG. 1.

In order to fully understand the configuration and effects of the present invention, preferred embodiments of the present invention will be described with reference to the attached drawings. However, the present invention is not limited to the embodiments disclosed below, but can be implemented in various forms and various changes can be made. However, the description of this embodiment is provided to ensure that the disclosure of the present invention is complete and to fully inform those skilled in the art of the present invention of the scope of the invention. In the attached drawings, components are shown enlarged in size for convenience of explanation, and the proportions of each component may be exaggerated or reduced.

Terms such as 'first' and 'second' may be used to describe various components, but the components should not be limited by the above terms. The above terms may be used only for the purpose of distinguishing one component from another. For example, the 'first component' may be named 'the second component' without departing from the scope of the present invention, and similarly, the 'second component' may also be named 'the first component'. You can. Additionally, singular expressions include plural expressions, unless the context clearly dictates otherwise. Unless otherwise defined, terms used in the embodiments of the present invention may be interpreted as meanings commonly known to those skilled in the art.

Below, with reference to the drawings, a blockchain-based copy prevention and deep learning AI-utilizing image processing voice/music output SVS system according to a preferred embodiment of the present invention will be described in detail. The photo interrupter will be described.

Figure 1 is a block diagram of an image processing voice/music output SVS system using blockchain-based copy prevention and deep learning AI according to an embodiment of the present invention.

Referring to FIG. 1, the present invention includes a service server 20 that receives subscription information from a plurality of member terminals 10, approves membership registration, and provides main services, and connects with the service server 20 to provide celebrity registration. A content production server 30 that produces a congratulatory song and provides voice data of celebrities, receives and stores the congratulatory song data and voice data from the content production server 30, and stores the congratulatory song data and voice data when requested by the service server 20. An additional service server 40 that provides a congratulatory service to the member terminal 10, learns the voice data, reproduces the name of the member to be provided with the additional service as the voice of the selected celebrity, and inserts it into a congratulatory song. It is configured to include a content processing server 50 provided by.

Hereinafter, the configuration and operation of the blockchain-based copy prevention and deep learning AI-utilizing image processing voice/music output SVS system according to a preferred embodiment of the present invention configured as described above will be described in more detail.

First, the member terminal 10 is equipped with a communication module capable of communication, includes a playback module capable of playing audio files or video files, and is preferably a computing device on which a dedicated application can be installed. For example, the member terminal 10 can be a smartphone, tablet, PC, etc.

The member terminal 10 is assumed to be able to use the service by accessing the web, mobile web, cloud, etc. provided by the service server 20.

The service server 20 may be an online shopping mall that sells physical goods, an online shopping mall that sells exchange vouchers such as Kifticon, or a card company or bank that issues cards and earns commission revenue when the cards are used.

The above example of the service server 20 is only for ease of understanding, and can be easily applied to a variety of fields, such as a securities company server and a game service server.

In addition, although the service server 20 is shown as a single configuration in the drawing, multiple service servers 20 providing different services may be applied. Specifically, the content production server 30 and the additional service server 40 can be simultaneously connected to a plurality of service servers 20 to provide content production and additional services.

In other words, the content creation server 30 and the additional service server 40 may be operated by an independent business operator separate from the service server 20.

The service server 20 can collect and store member information in response to a member registration request from the member terminal 10, and at this time, information on celebrities preferred by the member who owns the member terminal 10 is also collected.

Additionally, information on anniversaries for which members would like to be celebrated is provided and stored.

The service server 20 can collect information on preferred celebrities by providing a list of celebrities to the member terminal 10 and allowing the member to select a celebrity.

Celebrities can be set directly on the service server 20 or can be selected from the member terminal 10 by posting a list of celebrities provided by the content creation server 30.

The service server 20 can provide a service that approves membership registration of the member terminal 10 and accumulates points when using the service.

If the service server 20 is a shopping mall that sells products or exchange vouchers, the use of services for which points are accumulated can be considered as the purchase of goods or exchange vouchers, and if the service server 20 is a card company, card (credit or debit) You can accumulate points based on your usage performance.

At this time, the card may include an app card and a physical card, and the app card and physical card may be a rechargeable prepaid card.

The service server 20 provides information on celebrities preferred by members to the content production server 30 in accordance with the member's consent to use the service.

The content production server 30 may be operated by a celebrity agency or video producer. The content production server 30 may produce and store in advance various congratulatory songs that match the congratulatory content.

In other words, you can create a congratulatory song that meets the various needs of various members, such as a member's birthday, wedding anniversary, promotion date, discharge date, etc., and save it as a digital file. You can also create a video that matches the congratulatory song and save it as a digital file.

At this time, the congratulatory song can be produced with a voice suitable for the lyrics. The sound data of the congratulatory video can use the produced congratulatory song.

Alternatively, the content production server 30 can have a celebrity sing a congratulatory song and record it to produce a congratulatory song and a congratulatory video.

At this time, when playing later, the part where the member's name will be entered is recorded by pronouncing the promised syllable. For example, the part to be replaced with a name, such as 'ding-ding-ding-ding', is pronounced as a different syllable so that it can be clearly recognized.

Additionally, the content production server 30 produces voice data files recorded by celebrities.

Celebrities' voice data files can be created by having celebrities read the same script and save it as digital data.

The script consists of multiple sentences, and each of the multiple sentences can be read repeatedly a set number of times and recorded simultaneously.

This can be understood as an attempt to classify syllables with accurate pronunciation in the subsequent data learning process.

The content created in the content creation server 30 is provided to the content processing server 50, and then the content is processed and provided according to the request of the additional service server 40.

Figure 2 is a block diagram of one implementation of the content processing server 50.

Referring to FIG. 2, the content processing server 50 may include a storage unit 51, a learning unit 52, a syllable confirmation unit 53, a pitch confirmation unit 54, and a data processing unit 55.

The content created by the content creation server 30 is stored in the storage unit 51 of the content processing server 50.

The learning unit 52 of the content processing server 50 learns the celebrity's voice data file stored in the storage unit 51, classifies the syllables pronounced by the celebrity, and stores them in the storage unit 51.

This can be understood as the process of identifying and extracting the characteristics of a celebrity's voice for each syllable when reading a sentence written in a script.

In this way, repeatedly pronounced syllables can be collected and classified from the voice data file of a famous person, and the voice data of representative syllables with characteristic features can be individually stored.

The representative syllable voice data of celebrities stored in this way may be provided to the additional service server 40. This suggests that the subject of processing may be the additional service server 40.

Thereafter, when a request from the member terminal 10 or a set anniversary arrives, the service server 20 requests the additional service from the additional service server 40.

At this time, requests for additional services may include information on preferred celebrities, types and dates of anniversaries such as birthdays, member's age information, gender information, reception information, and member name.

The additional service server 40 provides information on celebrities, types of anniversaries such as birthdays, member age, gender information, and member names to the content processing server 50.

The anniversary date and reception information are used only when the additional service server 40 provides additional services, and are not provided to the content processing server 50.

The content processing server 50 selects the congratulatory song or video data stored in the storage unit 51 according to the type of anniversary received. There may be multiple congratulatory songs or videos depending on the type of anniversary.

At this time, the content processing server 50 checks whether there is a congratulatory song or video produced for the currently famous celebrity among the congratulatory songs or videos according to the type of anniversary.

If there is a song or video produced by the celebrity, select it. If not, select a song or video appropriate for the member's age and gender. At this time, the celebratory song data and video data may be tagged with appropriate age and gender information.

In this way, the representative syllable voice data of the preferred celebrity is inserted into the sound data of the celebratory song data or video data in the selected song data or video data.

Specifically, the selected congratulatory song data includes a data portion to be replaced with the member's name as described above, and the syllable confirmation unit 53 of the content processing server 50 determines the data promised in the selected congratulatory song data or the audio data of the video. Check if there is

At this time, considering that most Korean people's names have three syllables, the promised data can be recorded as data corresponding to three syllables.

That is, after finding three consecutive syllables of data and confirming it as the part where the member's name is to be replaced, the data processing unit 55 extracts voice data corresponding to the member's name from the representative syllable voice data and replaces the promised data.

At the same time, the pitch confirmation unit 54 checks the pitch information of the data to be replaced with the member's name, and when the promised data is replaced, the data processing unit 55 adjusts the pitch of the voice data corresponding to the member's name and then replaces it. .

At this time, if the member's name does not have three syllables, the data processing unit 55 can adjust the length of the extracted representative syllable voice data and replace it.

If a member's name has two syllables, the playback time of the syllable voice data is delayed and processed to fit a celebratory song. Conversely, if it has more than 4 syllables, the playback time is shortened and inserted into a congratulatory song or video.

In this way, the congratulatory song or video is processed by the content processing server 50, and is provided to the additional service server 40.

The additional service server 40 can provide the created congratulatory song or video to the member terminal 10 in various ways. For example, congratulatory songs or videos can be provided in a variety of ways, such as messages, chat apps, direct messages, downloads from the web, or push notifications from dedicated apps. At this time, the member's accumulated points may be deducted.

However, the provided data can be verified as original and cannot be copied or processed.

For example, a celebratory song or video may be processed into a non-fungible token (NFT).

In this way, the present invention can provide an additional service that provides all or part of a congratulatory song to members who use the service using voice data of celebrities, and the use of the main service can be further activated according to the provision of this additional service. There will be.

Although embodiments according to the present invention have been described above, they are merely illustrative, and those skilled in the art will understand that various modifications and equivalent scope of embodiments are possible therefrom. Therefore, the true technical protection scope of the present invention should be determined by the following claims.

10: Member terminal 20: Service server
30: Content production server 40: Additional service server
50: Content processing server 51: Storage unit
52: Learning section 53: Syllable confirmation section
54: Pitch confirmation unit 55: Data processing unit

Claims (7)

A service server that provides main services to members registered through member terminals;
A content production server that produces and stores the produced congratulatory song or video content as well as the voice data of celebrities;
a content processing server that extracts a syllable voice from the voice data and replaces the member's name part of the congratulatory song or video content with a syllable voice; and
When there is a request from the service server, it includes an additional service server that provides a congratulatory song or video processed by the content processing server to the member terminal,
The content processing server,
a storage unit that stores content created in the content creation server;
a learning unit that learns content that is voice data of a celebrity stored in the storage unit, collects and classifies syllables pronounced by the celebrity, and stores voice data of representative syllables in which characteristics appear in the storage unit;
A syllable confirmation unit that checks whether there is a data portion to be replaced with the member's name in the congratulatory song or video content;
a data processing unit that replaces the syllable of the celebrity corresponding to the member's name in the data to be replaced confirmed by the syllable confirmation unit;
It includes a pitch confirmation unit that adjusts the pitch of the celebratory song or video content to replace the syllables of the celebrity corresponding to the member's name,
A system that adjusts the playback time when a member's name does not have three syllables. According to paragraph 1,
The system is characterized in that the voice data of the celebrity is stored as digital data as the voice of the celebrity repeatedly reads a script containing a plurality of sentences. According to paragraph 2,
The content processing server,
A system that divides the voice data into syllables, classifies the voice of each syllable through artificial intelligence learning, and then determines a characteristic representative syllable voice. According to paragraph 1,
The congratulatory song or video content is,
Songs of celebrities or ordinary people are stored as digital data,
A system in which the parts containing the member's name in the congratulatory song are distinguished by repetition of promised syllables. According to paragraph 1,
The additional service server is,
A system characterized by processing congratulatory song or video content into a non-fungible token and providing it to member terminals. According to clause 5,
The additional service server is,
A system characterized by providing congratulatory song or video content to member terminals using app push, direct messaging, or messaging, or providing it to be downloaded from a service page provided by an additional service server. According to paragraph 1,
The service server is,
A system characterized by being an online shopping mall server, an exchange voucher shopping mall server, a credit card company server, or a bank server.