Video copywriting processing method and device, electronic equipment and storage medium

Abstract

The invention discloses a video copywriting processing method and apparatus, an electronic device and a storage medium. The method comprises the steps of obtaining a to-be-processed video polyphonic ringtone; performing text conversion processing on the to-be-processed video polyphonic ringtone to obtain a voice text data set; performing scene information analysis processing on the to-be-processed video polyphonic ringtone to obtain a video scene data set; performing industry feature recognition processing on the video scene data set to obtain industry feature information; and performing text generation processing on the voice text data set according to the video scene data set and the industry feature information to obtain a target video copywriting. According to the embodiment of the invention, personalized video copywriting can be generated, and the method can be widely applied to the technical field of computers.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a video text processing method, device, electronic device and storage medium. Background Art

[0002] Video ringback tone refers to the video content that users see before the call is connected when making a voice call or video call. In the related art, there is a way to generate video ringback tone through an agent channel, that is, to modify key information using a general text solution and a video template to obtain a video ringback tone. However, in actual applications, it is found that the user groups in different industries vary greatly, and the general text solution cannot meet the needs of different industries, resulting in the text and video materials being out of sync, the text content is not smooth enough, and the promotional content is not comprehensive enough, which affects the user experience. In summary, the technical problems existing in the related technology need to be improved. Summary of the invention

[0003] The main purpose of the embodiments of the present application is to provide a video text processing method, device, electronic device and storage medium, which can improve the user experience.

[0004] To achieve the above purpose, one aspect of an embodiment of the present application provides a video copy processing method, the method comprising:

[0005] Get the video ringback tone to be processed;

[0006] Performing text conversion processing on the video ringback tone to be processed to obtain a speech text data set;

[0007] Performing scene information analysis on the video ringback tone to be processed to obtain a video scene data set;

[0008] Performing industry feature recognition processing on the video scene data set to obtain industry feature information;

[0009] The speech text data set is processed for text generation according to the video scene data set and the industry characteristic information to obtain a target video copy.

[0010] In some embodiments, the step of converting the video ringback tone to be processed into text to obtain a speech text data set comprises the following steps:

[0011] Performing background music filtering processing on the voice file of the video ringback tone to be processed to obtain text dry sound;

[0012] The dry sound of the text is processed by speech recognition to obtain the speech text data set.

[0013] In some embodiments, the step of performing scene information analysis on the video ringback tone to be processed to obtain a video scene data set includes the following steps:

[0014] Performing video object and character action recognition processing on the video ringback tone to be processed to obtain a first scene data set;

[0015] Perform scene time span and screen dwell length recognition processing on the to-be-processed video ringback tone to obtain a second scene data set;

[0016] The video scene data set is obtained according to the first scene data set and the second scene data set.

[0017] In some embodiments, performing industry feature recognition processing on the video scene data set to obtain industry feature information includes the following steps:

[0018] Performing industry label classification processing on the video scene data set through a generative pre-training model to obtain main industry label data;

[0019] Subdividing the primary industry label data into grades and types to obtain secondary industry label data;

[0020] The industry characteristic information is obtained according to the primary industry label data and the secondary industry label data.

[0021] In some embodiments, performing text generation processing on the speech text data set according to the video scene data set and the industry characteristic information to obtain a target video copy includes the following steps:

[0022] Performing scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene text set;

[0023] The speech text data set is subjected to natural language optimization processing according to the video scene text set to obtain the target video text.

[0024] In some embodiments, performing scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene text set includes the following steps:

[0025] Determine primary industry label data and secondary industry label data according to the industry characteristic information;

[0026] Determine an industry copywriting factor set according to the primary industry label data and the secondary industry label data;

[0027] Determine a first scene data set and a second scene data set according to the video scene data set;

[0028] Performing similarity calculation processing on the industry copywriting factor set according to the first scene data set to obtain video scene information;

[0029] The video scene information is segmented according to the second scene data set to obtain the video scene text set.

[0030] In some embodiments, after obtaining the target video copy, the method further includes the following steps:

[0031] Perform background music matching processing on the target video copy according to the industry characteristic information to obtain the industry video background music;

[0032] Performing speech synthesis processing on the industry video background sound to obtain a speech file;

[0033] The video file of the to-be-processed video ringback tone and the voice file are synthesized to obtain a target video ringback tone.

[0034] To achieve the above-mentioned purpose, another aspect of the embodiment of the present application provides a video copy processing device, the device comprising:

[0035] The first module is used to obtain the video ringback tone to be processed;

[0036] The second module is used to perform text conversion processing on the video ringback tone to be processed to obtain a speech text data set;

[0037] The third module is used to analyze and process the scene information of the video ringback tone to be processed to obtain a video scene data set;

[0038] The fourth module is used to perform industry feature recognition processing on the video scene data set to obtain industry feature information;

[0039] The fifth module is used to perform text generation processing on the speech text data set according to the video scene data set and the industry characteristic information to obtain a target video copy.

[0040] In some embodiments, the second module is used to perform text conversion processing on the video ringback tone to be processed to obtain a speech text data set, including:

[0041] The first unit is used to perform background music filtering on the voice file of the video ringback tone to be processed to obtain text dry sound;

[0042] The second unit is used to perform speech recognition processing on the dry sound of the text to obtain the speech text data set.

[0043] In some embodiments, the third module is used to perform scene information analysis on the video ringback tone to be processed to obtain a video scene data set, including:

[0044] The third unit is used to perform video object and character action recognition processing on the video ringback tone to be processed to obtain a first scene data set;

[0045] The fourth unit is used to perform scene time span and screen stay length recognition processing on the video ringback tone to be processed to obtain a second scene data set;

[0046] The fifth unit is configured to obtain the video scene data set according to the first scene data set and the second scene data set.

[0047] In some embodiments, the fourth module is used to perform industry feature recognition processing on the video scene data set to obtain industry feature information, including:

[0048] A sixth unit is used to perform industry label classification processing on the video scene data set through a generative pre-training model to obtain main industry label data;

[0049] The seventh unit is used to perform grade and type subdivision processing on the primary industry label data to obtain secondary industry label data;

[0050] The eighth unit is used to obtain the industry characteristic information according to the primary industry label data and the secondary industry label data.

[0051] In some embodiments, the fifth module is used to perform text generation processing on the speech text data set according to the video scene data set and the industry characteristic information to obtain a target video copy, including:

[0052] A ninth unit is used to perform scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene text set;

[0053] The tenth unit is used to perform natural language optimization processing on the speech text data set according to the video scene text set to obtain the target video text.

[0054] In some embodiments, the ninth unit is used to perform scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene text set, including:

[0055] A first subunit is used to determine primary industry label data and secondary industry label data according to the industry characteristic information;

[0056] A second subunit is used to determine an industry copywriting factor set according to the primary industry label data and the secondary industry label data;

[0057] A third subunit, configured to determine a first scene data set and a second scene data set according to the video scene data set;

[0058] A fourth subunit is used to perform similarity calculation processing on the industry copywriting factor set according to the first scene data set to obtain video scene information;

[0059] The fifth subunit is used to perform segmentation processing on the video scene information according to the second scene data set to obtain the video scene text set.

[0060] In some embodiments, the apparatus further comprises a sixth module, comprising:

[0061] The eleventh unit is used to perform background music matching processing on the target video copy according to the industry characteristic information to obtain the industry video background music;

[0062] The twelfth unit is used to perform speech synthesis processing on the background sound of the industry video to obtain a speech file;

[0063] The thirteenth unit is used to synthesize the video file of the to-be-processed video ringback tone and the voice file to obtain a target video ringback tone.

[0064] To achieve the above-mentioned purpose, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above-mentioned method when executing the computer program.

[0065] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0066] The embodiments of the present application include at least the following beneficial effects: The present application provides a video text processing method, device, electronic device and storage medium. The solution can analyze the video scene content by performing scene information analysis on the processed video ringback tone to obtain a video scene data set, facilitate the subsequent generation of video text with data such as industry tags, and improve the accuracy of video text generation. In addition, the solution performs industry feature recognition processing on the video scene data set to obtain industry feature information, and can perform industry classification and association processing on the processed video ringback tone, making the industry tag data more complete, and improving the efficiency of video text generation. In addition, the solution performs text generation processing on the voice text data set according to the video scene data set and industry feature information to obtain the target video text, and can generate video text that conforms to the merchant industry and is close to the video scene, thereby improving the video promotion effect and improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 is a flow chart of a video copy processing method provided by an embodiment of the present application;

[0068] Figure 2 This is a schematic diagram of industry type distribution provided by an embodiment of the present application;

[0069] Figure 3 yes Figure 1 Flow chart of step S102 in FIG.

[0070] Figure 4 It is a schematic diagram of a framework of a recognition model provided in an embodiment of the present application;

[0071] Figure 5 yes Figure 1 Flow chart of step S103 in FIG.

[0072] Figure 6 yes Figure 1 Flow chart of step S104 in FIG.

[0073] Figure 7 It is a schematic diagram of calculating the similarity between industry copywriting and video scenes provided in an embodiment of the present application;

[0074] Figure 8 is a flow chart of regenerating a video provided by an embodiment of the present application;

[0075] Fig. 9 It is a structural schematic diagram of a video copy processing device provided in an embodiment of the present application;

[0076] Fig.10 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0077] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0078] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0079] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0080] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0081] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0082] 1) Video ringback tone: When users make a voice call or video call on the VoLTE network, they will see a video content before the call is connected. Video ringback tone users can create or upload personalized video content or choose from the operator's video library. Different video content can also be set for different callers.

[0083] 2) Wav2Vec 2.0 model: It is a model that uses self-supervised learning to extract information from audio. This model can better understand the contextual information in the audio and provide rich audio feature representation. It can be applied to various fields such as language learning, speech analysis, and audio indexing.

[0084] 3) PyTorchVideo model: It is a deep learning library focused on video understanding. The PyTorchVideo model provides reusable, modular, and efficient components needed to accelerate video understanding research, supports different deep learning video components such as video models, video datasets, and video-specific transformations, and can achieve video action recognition, video classification, video augmentation, video object recognition, and other fields.

[0085] 4) Generative Pre-trained Transformer (GPT): Also known as the big text model, it is a powerful language model that is trained on a large amount of text data to generate high-quality, coherent and contextual text. GPT has a wide range of applications, from automatic content generation, dialogue systems to code generation.

[0086] In the related art, there are two main ways to generate video ringback tones for business customers. One is to generate it through an agent channel, using a general text solution and video template, and only modifying key information such as address, merchant name, etc. to obtain a video ringback tones. The other is for the merchant to directly provide the video. However, due to the great differences in user groups in different industries, it is necessary to design copywriting based on the characteristics of different industries to attract user attention and thus trigger demand, and the general text solution cannot meet the needs of different industries. In addition, although customers understand the nature of their own industry, they lack professional copywriting capabilities, which may cause the video to fail to correctly reflect the characteristics of the industry, resulting in the copywriting and video materials being out of sync, the copywriting content being not smooth enough, and the promotional content being not comprehensive enough.

[0087] In view of this, a video text processing method, device, electronic device and storage medium are provided in the embodiments of the present application. The scheme obtains a video scene data set by performing scene information analysis on the processed video ringback tone, and can analyze the video scene content, facilitate the subsequent generation of video text with data such as industry labels, and improve the accuracy of video text generation. In addition, the scheme performs industry feature recognition processing on the video scene data set to obtain industry feature information, and can perform industry classification and association processing on the processed video ringback tone, making the industry label data more complete, and improving the efficiency of video text generation. In addition, the scheme performs text generation processing on the voice text data set according to the video scene data set and industry feature information to obtain the target video text, and can generate video text that conforms to the merchant industry and is close to the video scene, thereby improving the video promotion effect and improving the user experience.

[0088] A video copy processing method provided in an embodiment of the present application relates to the field of computer technology, can improve the quality of video ringback tone, highlight the personality and characteristics of customers, and thus improve the video promotion effect. The solution uses speech-to-text recognition technology and a large language model to extract a speech-to-text data set from a video file, and can extract merchant introduction information; and with the help of the PyTorchVideo tool model, the video scene information is analyzed. By integrating video scene information, customer industry characteristics and speech-to-text content, the Wenshengwen large model is used to optimize the merchant introduction copy content. The solution can save labor costs, improve efficiency, enhance copy quality, and can be applied to industries such as catering, retail, and tourism to help customers improve video promotion effects. A video copy processing method provided in an embodiment of the present application can be applied to a terminal, can also be applied to a server, and can also be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a car terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or it can be configured as a server cluster or distributed system composed of multiple physical servers. It can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in the blockchain network; the software can be an application that implements a video copy processing method, etc., but is not limited to the above forms.

[0089] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0090] It should be noted that in each specific implementation of the present application, when it comes to the need to perform relevant processing based on data related to user identity or characteristics such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of these data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0091] Figure 1 is an optional flowchart of a video copy processing method provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S105.

[0092] Step S101, obtaining a video ringback tone to be processed;

[0093] Step S102, performing text conversion processing on the video ringback tone to be processed to obtain a speech text data set;

[0094] Step S103, performing scene information analysis on the video ringback tone to be processed to obtain a video scene data set;

[0095] Step S104, performing industry feature recognition processing on the video scene data set to obtain industry feature information;

[0096] Step S105 , performing text generation processing on the speech text data set according to the video scene data set and the industry characteristic information to obtain a target video copy.

[0097] In the steps S101 to S105 shown in the embodiment of the present application, a speech-text data set is obtained by converting the video ringback tone to be processed into text, thereby optimizing the original video text to obtain the target video copy, so as to improve the quality of the video ringback tone. Specifically, the ringtone content of the video ringback tone to be processed is converted into text data and key information is extracted to obtain the speech-text data set. Then, the scene information of the video ringback tone to be processed is analyzed and processed, and the information such as objects, actions, people, and light in the scene is obtained by identifying the picture of the video ringback tone, and the length of time when the object action appears is recorded to determine the video scene data set. Please refer to Figure 2 , Figure 2It is a schematic diagram of the industry type distribution provided by the embodiment of the present application, which shows the type distribution of video ringback tone in various industries, including IT Internet, personal occupation, transportation, leisure and entertainment and other industries. Since the user groups in different industries vary greatly, it is necessary to design copywriting according to the characteristics of different industries. Therefore, the embodiment of the present application performs industry feature recognition processing based on the video scene data set to obtain industry feature information, which can identify the main industry type of the customer and the probability of related industries, and optimize the video copywriting through the industry feature information obtained by recognition. Based on the above information, text generation processing can be performed through the Wenshengwen model to regenerate a new solution text and obtain the target video copy. The embodiment of the present application optimizes the content of the video copywriting through speech recognition, text generation and other processing, which can save labor costs, improve efficiency and enhance the quality of the copywriting.

[0098] In step S101 of some embodiments, the video ringback tone to be processed may be obtained by a user uploading a video ringback tone, or by other means, such as obtaining a video from a video library as the video ringback tone to be processed, but the present invention is not limited thereto.

[0099] See also Figure 3 In step S102 of some embodiments, the text conversion process is performed on the video ringback tone to be processed to obtain a speech text data set, including the following steps:

[0100] Step S301, performing background music filtering processing on the voice file of the video ringback tone to be processed to obtain text dry sound;

[0101] Step S302, performing speech recognition processing on the dry sound of the text to obtain the speech text data set.

[0102] In the embodiment of the present application, by reading the voice file of the video ringback tone to be processed and converting it into text, it is possible to analyze the customer's product information, business scope and industry field to which it belongs, and obtain a voice-to-text data set. This information can provide a high reference value for the optimization of the later voice solution and language modification. Since the voice file of the video ringback tone to be processed usually contains two parts: the dry sound of the text and the background music, in order to avoid the interference of the background music on the voice analysis, the embodiment of the present application will remove the background music and only use the dry sound part of the text for analysis. Please refer to Figure 4In the embodiment of the present application, the optimized wav2vec 2.0 speech recognition model is used to perform speech recognition processing on the dry voice of the text to obtain the speech text data set, and the original speech waveform is input into the speech recognition model, and the convolution layer (CNN) is used to process the waveform to obtain the quantized representation and the hidden layer representation, and then the mask module and the transformer module are used to process the waveform to obtain the context representation, and the representation is learned by maximizing the similarity between positive examples and minimizing the similarity between negative examples to obtain the final contrast loss. The embodiment of the present application specifically reads the dry text part of the video ringtone and extracts the nouns therein, thereby obtaining the key content of the customer ringtone. These key contents will be used for the video industry analysis and text copy optimization in the next step. At the same time, the embodiment of the present application will save special key information such as telephone number, merchant address, merchant name, main products or business, etc. from the identified key content to form a voice text data set. The embodiment of the present application can reflect the processed voice text data set in the subsequent optimization process by performing text conversion processing on the processed video ringtone to ensure the integrity and effectiveness of the video information.

[0103] See also Figure 5 In step S103 of some embodiments, the performing scene information analysis on the to-be-processed video ringback tone to obtain a video scene data set includes the following steps:

[0104] Step S501, performing video object and character action recognition processing on the video ringback tone to be processed to obtain a first scene data set;

[0105] Step S502, performing scene time span and screen stay length recognition processing on the to-be-processed video ringback tone to obtain a second scene data set;

[0106] Step S503: Obtain the video scene data set according to the first scene data set and the second scene data set.

[0107] In an embodiment of the present application, the video ringback tone to be processed is input into the PyTorchVideo tool model, which obtains the scene information and spatial temporality of the video by understanding the calculation to obtain the video scene data set. Specifically, by identifying the light, objects, people and actions in the video screen, and generating a series list containing specific items and action names, the data is saved as the first scene data set. For example, if the video screen appears buildings, rooms, sheets, room items, room lighting atmosphere and the final sales information in sequence, the embodiment of the present application will record the continuous changes of the scene and the specific content of each object, and save it as the first scene data set, wherein the first scene data set mainly saves the objects, character action events, etc. that appear in the video scene, and can correctly reflect the video ownership, information, promotional content, and specific things through objects and actions. Then the scene time span and screen stay length recognition processing are performed on the video ringback tone to be processed, by recording the time span and screen stay length of each scene in the video screen. Different time lengths can reflect which content is highlighted and help determine the font length required for the promotional copy. For example, the video screen shows buildings (3 seconds) -> rooms (8 seconds) -> bed sheets (3 seconds) -> items in the room (5 seconds) -> room lighting atmosphere (2 seconds) -> the last sale (4 seconds). The embodiment of the present application uses these time information as the time attribute of the scene and saves it to the second scene data set. The second scene data set saves the time length of each type of object action. The content that needs to be promoted in the video focus, the frequency of screen switching, etc. can be determined through length analysis. The embodiment of the present application can provide a data basis for later video analysis by performing scene information analysis on the processed video ringback tone.

[0108] See also Figure 6 In step S104 of some embodiments, performing industry feature recognition processing on the video scene data set to obtain industry feature information includes the following steps:

[0109] Step S601, performing industry label classification processing on the video scene data set through a generative pre-training model to obtain main industry label data;

[0110] Step S602, performing level and type subdivision processing on the primary industry label data to obtain secondary industry label data;

[0111] Step S603, obtaining the industry characteristic information according to the primary industry label data and the secondary industry label data.

[0112] In an embodiment of the present application, the industry label classification processing of the video scene data set is performed by a generative pre-training model to obtain the main industry label data. The generative pre-training model is a pre-trained GPT model. The training method can be trained by a large number of industry feature recognition and manual error correction methods, and the main training direction can be set to the sales industry. Since each industry will have its own main features and secondary features, the embodiment of the present application can calculate the industry probability by comparing the main features and combining the secondary features. For example, if the bed sheet content appears in the video, it can be preliminarily determined as a hotel, a bedding store, a large supermarket, and a real estate; then compare the specific content next to the bed sheet, such as the corridor, the bathroom, etc., and further increase the probability of the hotel, and exclude the bedding store and the large supermarket. The embodiment of the present application can calculate the label classification through the GPT model, and the output industry label division probability is: 70% hotel industry, 20% clothing industry, 10% construction industry. The embodiment of the present application can be calculated by comparing the lower limit of the threshold, that is, setting a threshold and comparing the maximum probability value. For example, if the threshold is set to 60%, the hotel industry can be determined as the main industry label of the customer, and other probabilities can be used as related industries. The embodiment of the present application can also perform grade and type segmentation processing on the main industry label data. According to the main industry information of the previous step, it can be further segmented to segment the grade and type of the same industry. For example, the hotel industry can be segmented into economy hotels, business hotels, tourist hotels, hot spring hotels and theme hotels. According to the combination of voice text information and video scene information, and then calculated by the GPT model, the specific segmentation type probability is obtained by comparing the segmentation characteristics of the same industry: 80% business hotels, 12% tourist hotels, 8% theme hotels, thereby determining business hotels as the secondary industry information of customers, and other probabilities as related industries. The embodiment of the present application obtains industry feature information by performing industry feature recognition processing on the video scene data set, which can determine the industry type involved in the business scope of the video ringback tone to be processed, and can provide a basis for the subsequent calculation of the customer group that the industry mainly faces. In this way, it can ensure that the customer's industry information and areas of involvement are accurate and comprehensive, thereby generating more accurate video copywriting and avoiding omissions.

[0113] In step S105 of some embodiments, performing text generation processing on the speech text data set according to the video scene data set and the industry characteristic information to obtain a target video copy includes the following steps:

[0114] Performing scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene text set;

[0115] The speech text data set is subjected to natural language optimization processing according to the video scene text set to obtain the target video text.

[0116] In the embodiment of the present application, because the audience groups of video ringback tone vary from industry to industry, the income level, education level and hobby types of the audience groups are different, so different copy content needs to be formulated for different industries. In addition, the product and service types of various industries are also different, and professional vocabulary is needed to accurately introduce the products and services in the video ringback tone. Therefore, it is necessary to select different vocabularies according to different industries, and perform scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene copy set. Then, the video scene copy set is optimized by verb, adjective modification, vocabulary replacement and other optimization processing, and the copy length can also be limited to not exceed the word limit. At the same time, the voice text data set is added for natural language optimization processing, and the position of the third data set data in the newly generated text scheme scene can be kept unchanged to generate new copy content. The embodiment of the present application can perform natural language optimization through the GPT model, making the text more natural, beautiful and smooth while keeping the overall font number and length unchanged. The embodiment of the present application obtains the target video copy by performing text generation processing on the voice text data set according to the video scene data set and the industry characteristic information, which can cover all the information of the service content, highlight the characteristic service, and improve the generation efficiency and publicity effect of the video copy.

[0117] In some embodiments, performing scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene text set includes the following steps:

[0118] Determine primary industry label data and secondary industry label data according to the industry characteristic information;

[0119] Determine an industry copywriting factor set according to the primary industry label data and the secondary industry label data;

[0120] Determine a first scene data set and a second scene data set according to the video scene data set;

[0121] Performing similarity calculation processing on the industry copywriting factor set according to the first scene data set to obtain video scene information;

[0122] The video scene information is segmented according to the second scene data set to obtain the video scene text set.

[0123] In the embodiment of the present application, the primary industry label data and the secondary industry label data are obtained from the industry characteristic information. By selecting a primary industry label data, a special copywriting factor set with the primary industry type as the industry can be screened out. For example, based on the hotel industry, a special copywriting factor set including copywriting words such as clean, grand, and cost-effective can be obtained. By selecting a secondary industry label data, a special copywriting factor set with the secondary industry type as the industry is screened out, and an industry copywriting factor set is obtained based on the combination of the primary industry label data and the secondary industry label data. The first scene data set and the second scene data set can be obtained from the video scene data set. Please refer to Figure 7 , taking the industry copywriting factor set as the x-axis and the first scene data set as the y-axis, constructing the user similarity calculation coordinate system, and calculating the similarity of the two sets of feature factors according to the discrete distribution of the data on the x-axis and y-axis coordinate systems, respectively. According to the similarity value comparison, the higher the similarity, the more the copywriting in the industry copywriting factor set conforms to the video scene information V(X). The expression of the video scene information is as follows:

[0124]

[0125] In the formula, A represents any copy factor in the industry copy factor set, and B represents the scene data factor in the first scene data set. In the embodiment of the present application, the copy with the highest calculated video scene information is used as the key copy content of the video scene, and then the key video copy content is cut into one-sentence promotional content of different lengths according to the main objects and time span that appear in the second scene data set. The number of words in the content is determined by the time span in which the main objects appear, and the video scene copy set with each main promotional content and word limit is calculated. In the embodiment of the present application, the video scene data set is subjected to scene text generation processing according to industry characteristic information to obtain a video scene copy set, and personalized text solutions can be formulated for different industries and different video contents, so as to more effectively promote the service content, service scope, service features and merchant advantages of video ringback tones.

[0126] In some embodiments, after obtaining the target video copy, the method further includes the following steps:

[0127] Perform background music matching processing on the target video copy according to the industry characteristic information to obtain the industry video background music;

[0128] Performing speech synthesis processing on the industry video background sound to obtain a speech file;

[0129] The video file of the to-be-processed video ringback tone and the voice file are synthesized to obtain a target video ringback tone.

[0130] In an embodiment of the present application, the target video text can be matched with background music according to the industry characteristic information, and the industry video background sound can be selected. In an embodiment of the present application, a suitable professional voice anchor can be selected from the anchor library, and a voice file can be generated by multi-timbre voice synthesis technology. Finally, the video file of the processed video ringback tone and the newly synthesized voice text are synthesized and a new video file is regenerated, and a new video ringback tone can be loaded to complete the update of the video ringback tone. The embodiment of the present application can generate a new video ringback tone, which can improve the personalized display of the video ringback tone and enhance the user experience.

[0131] The following is a detailed description of the embodiments of the present invention with reference to specific application examples:

[0132] The embodiments of the present application can be applied to the computer field, and can automatically generate introduction texts of videos or video ringback rings for users, and can also automatically generate video ringback rings based on video texts. Figure 8 In the embodiment of the present application, the ringtone file provided by the customer is converted into text to extract the ringtone text content, and based on the video ringtone uploaded by the customer, the PyTorchVi deo tool is used to obtain the customer's video scene information and the industry to which it belongs. Then, the video scene and the ringtone text content are combined with the industry characteristic information to obtain the customer's industry copy content. Finally, the optimized text copy is selected by a suitable anchor for voice generation, and the audio file is synthesized to regenerate the video. The embodiment of the present application can effectively mine promotional copy that is in line with the merchant industry and close to the video scene, and generate personalized video copy by covering industry characteristics through intelligent voice generation means, thereby reducing manual processing costs, improving customer promotion effects, and thus improving conversion rates.

[0133] See also Fig. 9 The embodiment of the present application also provides a video copy processing device, which can implement the above-mentioned video copy processing method, and the device includes:

[0134] The first module 901 is used to obtain a video ringback tone to be processed;

[0135] The second module 902 is used to perform text conversion processing on the video ringback tone to be processed to obtain a speech text data set;

[0136] The third module 903 is used to analyze and process the scene information of the video ringback tone to be processed to obtain a video scene data set;

[0137] The fourth module 904 is used to perform industry feature recognition processing on the video scene data set to obtain industry feature information;

[0138] The fifth module 905 is used to perform text generation processing on the speech text data set according to the video scene data set and the industry characteristic information to obtain a target video copy.

[0139] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0140] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned video copy processing method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0141] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0142] See also Fig.10 , Fig.10 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0143] The processor 1001 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0144] The memory 1002 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 1002 can store an operating system and other application programs. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 1002, and the processor 1001 calls and executes the video text processing method of the embodiment of this application;

[0145] Input / output interface 1003, used to implement information input and output;

[0146] The communication interface 1004 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WI FI, Bluetooth, etc.);

[0147] A bus 1005 , which transmits information between various components of the device (e.g., the processor 1001 , the memory 1002 , the input / output interface 1003 , and the communication interface 1004 );

[0148] The processor 1001 , the memory 1002 , the input / output interface 1003 and the communication interface 1004 are connected to each other in communication within the device via the bus 1005 .

[0149] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned video copy processing method is implemented.

[0150] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0151] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0152] The embodiments of the present application provide a video text processing method, device, electronic device and storage medium, which performs scene information analysis on the processed video ringback tone to obtain a video scene data set, can analyze the video scene content, facilitate the subsequent generation of video text with industry labels and other data, and improve the accuracy of video text generation. In addition, the scheme performs industry feature recognition processing on the video scene data set to obtain industry feature information, can perform industry classification and association processing on the processed video ringback tone, make the industry label data more complete, and improve the efficiency of video text generation. In addition, the scheme performs text generation processing on the voice text data set according to the video scene data set and industry feature information to obtain the target video text, and can generate video text that conforms to the merchant industry and is close to the video scene, thereby improving the video promotion effect and improving the user experience.

[0153] Based on the industry information of the video ringback tone, the embodiment of the present application constructs a model that distinguishes industry information characteristics and the target user group of the merchant. It can combine the scene information of the video uploaded by the customer, optimize the calculation of the video text copy, generate a voice file in the context of the integrated video, and recommend it to the customer. It solves the problem that the prior art cannot combine the video scene and industry characteristics for accurate copy generation and audio synthesis. By intelligently optimizing the voice copy, the embodiment of the present application can better introduce and promote customer products and services, and ensure the personalization, automation and rationalization of the generated copy. In addition, the embodiment of the present application can also effectively use the feedback information of other merchant industries to generate video ringback tones, speed up the effect of personalized generation, and improve the conversion rate.

[0154] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0155] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0156] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0157] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0158] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0159] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0160] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0161] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0162] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0163] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0164] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A video copy processing method, characterized in that: The method comprises the following steps: Get the video ringback tone to be processed; Performing text conversion processing on the video ringback tone to be processed to obtain a speech text data set; Performing scene information analysis on the video ringback tone to be processed to obtain a video scene data set; Performing industry feature recognition processing on the video scene data set to obtain industry feature information; The speech text data set is processed for text generation according to the video scene data set and the industry characteristic information to obtain a target video copy.

2. The method according to claim 1, characterized in that The step of converting the video ringback tone to be processed into text to obtain a speech text data set comprises the following steps: Performing background music filtering processing on the voice file of the video ringback tone to be processed to obtain text dry sound; The dry sound of the text is processed by speech recognition to obtain the speech text data set.

3. The method according to claim 1, characterized in that The step of performing scene information analysis on the video ringback tone to be processed to obtain a video scene data set comprises the following steps: Performing video object and character action recognition processing on the video ringback tone to be processed to obtain a first scene data set; Perform scene time span and screen dwell length recognition processing on the video ringback tone to be processed to obtain a second scene data set; The video scene data set is obtained according to the first scene data set and the second scene data set.

4. The method according to claim 1, characterized in that: The step of performing industry feature recognition processing on the video scene data set to obtain industry feature information comprises the following steps: Performing industry label classification processing on the video scene data set through a generative pre-training model to obtain main industry label data; The primary industry label data is subdivided into grades and types to obtain secondary industry label data; The industry characteristic information is obtained according to the primary industry label data and the secondary industry label data.

5. The method according to claim 1, characterized in that The step of performing text generation processing on the speech text data set according to the video scene data set and the industry characteristic information to obtain a target video copy includes the following steps: Performing scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene text set; The speech text data set is subjected to natural language optimization processing according to the video scene text set to obtain the target video text.

6. The method according to claim 5, characterized in that The step of performing scene text generation processing on the video scene data set according to the industry characteristic information to obtain a video scene text set includes the following steps: Determine primary industry label data and secondary industry label data according to the industry characteristic information; Determine an industry copywriting factor set according to the primary industry label data and the secondary industry label data; Determine a first scene data set and a second scene data set according to the video scene data set; Performing similarity calculation processing on the industry copywriting factor set according to the first scene data set to obtain video scene information; The video scene information is segmented according to the second scene data set to obtain the video scene text set.

7. The method according to any one of claims 1 to 6, characterized in that: After obtaining the target video copy, the method further includes the following steps: Perform background music matching processing on the target video copy according to the industry characteristic information to obtain the industry video background music; Performing speech synthesis processing on the industry video background sound to obtain a speech file; The video file of the to-be-processed video ringback tone and the voice file are synthesized to obtain a target video ringback tone.

8. A video copy processing device, characterized in that: The device comprises: The first module is used to obtain the video ringback tone to be processed; The second module is used to perform text conversion processing on the video ringback tone to be processed to obtain a speech text data set; The third module is used to analyze and process the scene information of the video ringback tone to be processed to obtain a video scene data set; The fourth module is used to perform industry feature recognition processing on the video scene data set to obtain industry feature information; The fifth module is used to perform text generation processing on the speech text data set according to the video scene data set and the industry characteristic information to obtain a target video copy.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images