Video push method, device and electronic equipment

Abstract

The embodiment of the present invention provides a video push method, device and electronic device, which relate to the field of data processing technology. The specific scheme is as follows: obtain the identifier and target trading day of the target financial product input by the user on the client; based on the identifier and target trading day, obtain the financial data of the target financial product; based on the financial data, generate a financial chart image of the target financial product; obtain news and public opinion data related to the target financial product; convert the news and public opinion data into text data in a preset data format; generate an audio segment of the voice that broadcasts the text data; based on the duration of the audio segment and the financial chart image, generate a video frame of the target video; synchronize the audio segment with the video frame of the target video to obtain the target video, and feed back the target video to the client. Applying the scheme provided by the embodiment of the present invention can improve the efficiency of users in obtaining financial information.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a video pushing method, device and electronic equipment. Background Art

[0002] In the financial sector, users rely on various types of financial information to understand market trends, company performance, policy changes, and more. This allows them to better assess the risks of financial products. Therefore, financial information plays a crucial role for users. However, access to financial information is complex, and the volume of unstructured information is substantial. Conventional methods typically rely on manual searches and subscriptions to numerous websites and news sources to obtain financial information, resulting in low efficiency. Summary of the Invention

[0003] The purpose of the embodiments of the present invention is to provide a video push method, device, and electronic device to improve the efficiency of users in obtaining financial information. The specific technical solution is as follows:

[0004] According to one aspect of an embodiment of the present invention, a video push method is provided, which is applied to a server, and the method includes:

[0005] Obtain the identifier of the target financial product and the target trading date entered by the user in the client;

[0006] Based on the identifier and the target trading date, obtaining financial data of the target financial product;

[0007] generating a financial chart image of the target financial product based on the financial data;

[0008] Obtain news and public opinion data related to the target financial product;

[0009] Converting the news and public opinion data into text data in a preset data format;

[0010] generating an audio segment of speech that reads the text data;

[0011] generating a video frame of a target video based on the duration of the audio segment and the financial chart image;

[0012] The audio segment is synchronized with the video frame of the target video to obtain the target video, and the target video is fed back to the client.

[0013] According to another aspect of an embodiment of the present invention, a video push method is provided, which is applied to a client, and the method includes:

[0014] Obtaining the identifier and target trading date of the target financial product input by the user;

[0015] Sending the identifier of the target financial product and the target trading date to the server, so that the server generates a target video according to any of the above-mentioned video push methods applied to the server, and feeds the target video back to the client;

[0016] The target video is received and displayed.

[0017] According to another aspect of an embodiment of the present invention, a video push device is provided, which is applied to a server, and the device includes:

[0018] A first target information obtaining module is used to obtain the identifier of the target financial product and the target trading date input by the user on the client;

[0019] a financial data acquisition module, configured to acquire financial data of the target financial product based on the identifier and the target trading date;

[0020] a financial chart image generation module, configured to generate a financial chart image of the target financial product based on the financial data;

[0021] A news and public opinion data acquisition module, used to obtain news and public opinion data related to the target financial product;

[0022] A text data conversion module, used to convert the news and public opinion data into text data in a preset data format;

[0023] An audio segment generating module, configured to generate an audio segment for broadcasting the text data;

[0024] a video frame generation module, configured to generate a video frame of a target video based on the duration of the audio segment and the financial chart image;

[0025] The target video feedback module is used to synchronize the audio segments with the video frames of the target video to obtain the target video, and feed back the target video to the client.

[0026] According to another aspect of an embodiment of the present invention, a video push device is provided, which is applied to a client, and the device includes:

[0027] A second target information obtaining module is used to obtain the identifier of the target financial product and the target trading date input by the user;

[0028] a target information sending module, configured to send the identifier of the target financial product and the target trading date to the server, so that the server generates a target video according to any of the aforementioned video push devices applied to the server, and feeds the target video back to the client;

[0029] The target video display module is used to receive the target video and display the target video.

[0030] According to another aspect of an embodiment of the present invention, there is provided an electronic device, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0031] Memory for storing computer programs;

[0032] The processor is configured to implement any of the above-mentioned video pushing methods when executing a program stored in the memory.

[0033] According to another aspect of the embodiments of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any of the above-mentioned video pushing methods is implemented.

[0034] According to another aspect of the embodiments of the present invention, a computer program product including instructions is provided, which, when executed on a computer, enables the computer to execute any of the above-mentioned video pushing methods.

[0035] Beneficial effects of the embodiments of the present invention:

[0036] The video push method provided by the embodiment of the present invention is that when a user wants to obtain the financial information of a target financial product on a target trading day, the user can input the identifier of the target financial product and the target trading day on the user side and send a request to the server side. The server side collects the financial data of the target financial product and the news and public opinion data, processes the financial data to generate financial chart information, generates audio of the voice that broadcasts the Chinese characters in the news and public opinion data, and generates a target video that can introduce the financial information of the target financial product based on the audio and financial chart information, and feeds the generated target video back to the user side. The processed financial information is displayed to the user in the form of a dynamic and visual video. In this way, the user can obtain the target video that introduces the financial information of the target financial product in a more convenient way, which enables the user to obtain the financial information of the target financial product more quickly and easily, thereby improving the efficiency of the user in obtaining financial information.

[0037] Of course, it is not necessary to achieve all of the advantages described above simultaneously in order to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0039] Figure 1 A schematic diagram of a video push method according to an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of a flow chart of another video push method provided by an embodiment of the present invention;

[0041] Figure 3a A schematic diagram of a first type of financial chart image provided by an embodiment of the present invention;

[0042] Figure 3b A schematic diagram of a second type of financial chart image provided by an embodiment of the present invention;

[0043] Figure 3c A schematic diagram of a third type of financial chart image provided by an embodiment of the present invention;

[0044] Figure 3d A schematic diagram of a fourth type of financial chart image provided by an embodiment of the present invention;

[0045] Figure 4 A schematic diagram of a flow chart of a method for generating a video frame provided by an embodiment of the present invention;

[0046] Figure 5 A flowchart of another video push method provided by an embodiment of the present invention;

[0047] Figure 6 A schematic structural diagram of a video push device provided by an embodiment of the present invention;

[0048] Figure 7 A schematic structural diagram of another video push device provided by an embodiment of the present invention;

[0049] Figure 8 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field based on the present invention are within the scope of protection of the present invention.

[0051] The following describes the background technology of the embodiments of the present invention.

[0052] In the financial sector, users rely on various types of financial information to understand market trends, company performance, policy changes, and more. This allows them to better assess the risks of financial products. Therefore, financial information plays a crucial role for users. However, users often need to manually search online for the information they need and subscribe to relevant websites and news sources to obtain financial information. Furthermore, the financial information they retrieve often consists of both structured and unstructured information, making it difficult for users to filter and analyze this vast amount of unstructured information. Furthermore, collecting relevant data for structured information often requires strong expertise in computer science and finance, making it difficult for ordinary users to collect and analyze the vast amounts of unstructured information. Furthermore, the information users obtain is often disorganized and unintuitive. Even if they do obtain financial information, they still struggle to quickly understand and evaluate it.

[0053] In one embodiment of the present invention, see Figure 1 A flow chart of a video push method is provided. The method is applied to a server and includes the following steps S101-S108.

[0054] Step S101: Obtain the identifier of the target financial product and the target trading date input by the user on the client.

[0055] Financial products may include stocks, funds, futures, and bonds. If a user wishes to learn more about a target financial product, they can enter the target product's identifier on the client. The client then sends the identifier of the target financial product, whose information the user wishes to view, to the server, requesting a target video introducing the target financial product. Each financial product has a unique identifier. For example, if the target financial product is a stock, the identifier may be its stock ID (identification number). Alternatively, the user can enter the target trading day for which they wish to learn more about the target financial product. Alternatively, if the user does not enter a target trading day, and the current date is a non-trading day, the target trading day may be the most recent trading day before the date the user initiated the request for the target video introducing the target financial product using the client. If the current date is a trading day, the target trading day may be the date the user initiated the request for the target video introducing the target financial product using the client. In this way, the server can obtain the target financial product's identifier and target trading day by receiving the target financial product's identifier and target trading day entered by the user on the client.

[0056] Step S102: Based on the identifier and the target trading date, obtain financial data of the target financial product.

[0057] Step S103: Generate a financial chart image of the target financial product based on the financial data.

[0058] For different types of financial data, the server can obtain the financial data in different ways and then generate financial chart images of the target financial products in different ways. The way in which the server obtains the financial data and generates the financial chart images in steps S102-S103 is described in the following embodiments and will not be described in detail here.

[0059] Step S104: Obtain news and public opinion data related to the target financial product.

[0060] In one implementation, the server can determine the target company related to the target financial product based on the identifier of the target financial product, and then obtain news and public opinion data related to the target company within a second historical time from the data source of the news and public opinion data. For example, news content related to the target company, comments related to the target company in social media, etc. can be obtained. The second historical time is: a historical time of a preset time length before the target trading day, or a historical time of a preset time length before the current date. For example, the second historical time can be three months before the target trading day, or three months before the current date. Specifically, the server can obtain news and public opinion data through a preset news and public opinion data acquisition interface.

[0061] Step S105: converting the news and public opinion data into text data in a preset data format.

[0062] The preset data format may be a string array.

[0063] In one implementation, the server can directly convert the text of the obtained news and public opinion data into text data in a preset data format.

[0064] In another implementation, after obtaining the news and public opinion data, the server can sort the news and public opinion data in chronological order according to the time to which the news and public opinion data belongs, input the sorted news and public opinion data into a pre-trained large language model for summary extraction, obtain the news summary text of the news and public opinion data, and then convert the above news summary text into text data in a preset data format to obtain text data of the news summary text stored according to the preset data structure. The time to which the news and public opinion data belongs can be the time when the news and public opinion data is collected, or it can be the time when the news and public opinion data is released, and the large language model can be a large language model that has been tuned with data from financial vertical scenarios.

[0065] Other implementations of converting the data into text data in a preset data format in step S105 are described in the following embodiments and will not be described in detail here.

[0066] Step S106: Generate an audio segment of the voice that announces the text data.

[0067] The generated audio segments may be stored in the form of audio files. For example, the data format of the audio files may be WAV (WaveForm, audio file format).

[0068] In one implementation, the server may input the text data in the preset data format converted in step S105 into a preset speech generation model to generate an audio segment of the speech that reads the text data. The preset speech generation model may be one of the following: PnG BERT, Flow-TTS, Glow-TTS, DeepVoice3, and Tacontron2 (five audio generation models for text-to-speech).

[0069] Furthermore, when generating an audio segment for the voice broadcasting text data, the server may use pre-set default voice parameters to generate the audio segment. The voice parameters may include timbre and speech rate, and the default voice parameters may include a female voice with a neutral voice and a speech rate of 200 words per minute. The client may display a voice parameter selection button to the user on the user interface. The client may send the user-selected timbre and speech rate to the server, and the server may use the user-selected timbre and speech rate to generate the audio segment.

[0070] Other implementations of generating audio segments in step S106 are described in the following embodiments and are not described in detail here.

[0071] Step S107: Generate video frames of the target video based on the duration of the audio segment and the financial chart image.

[0072] In one implementation, the server may generate video frames that sequentially display each financial chart image for the duration of the audio segment as video frames of the target video.

[0073] Other implementations of generating the video frame of the target video in step S107 are described in the following embodiments and are not described in detail here.

[0074] Step S108: Synchronize the audio track of the audio segment with the video frame of the target video to obtain the target video, and feed the target video back to the client.

[0075] In one implementation, timestamp alignment is performed based on the timestamps of the audio segments and the timestamps of the video frames to synchronize the audio tracks of the audio segments with the video frames of the target video, thereby obtaining the target video. After obtaining the target video, the obtained target video is fed back to the client.

[0076] Other implementations of audio track synchronization in step S108 are described in the following embodiments and will not be described in detail here.

[0077] By applying the video push method provided by the embodiment of the present invention, when a user wants to obtain the financial information of a target financial product on a target trading day, the user can input the identifier of the target financial product and the target trading day on the user side and send a request to the server side. The server side collects the financial data of the target financial product and the news and public opinion data, processes the financial data to generate financial chart information, generates audio of the voice that broadcasts the Chinese characters in the news and public opinion data, and generates a target video that can introduce the financial information of the target financial product based on the audio and financial chart information, and feeds the generated target video back to the user side. By displaying the processed financial information to the user in the form of a dynamic and visual video, the user can obtain the target video that introduces the financial information of the target financial product in a more convenient way, which enables the user to obtain the financial information of the target financial product more quickly and easily, thereby improving the efficiency of the user in obtaining financial information.

[0078] In this way, users can quickly obtain intuitive and rich financial information without having relevant coding skills, lowering the technical threshold for users to obtain financial information.

[0079] In another embodiment of the present invention, see Figure 2 A flowchart of another video push method is provided, wherein the method includes the following steps S201-S211. The above steps S105-S106 can be implemented by the following steps S206-S207, the above step S107 can be implemented by the following steps S208-S210, and the above step S108 can be implemented by the following step S211.

[0080] Step S201: Obtain the identifier of the target financial product and the target trading date input by the user on the client.

[0081] Step S202: Based on the identifier and the target transaction date, obtain financial data of the target financial product.

[0082] Step S203: Generate a financial chart image of the target financial product based on the financial data.

[0083] Step S204: Obtain news and public opinion data related to the target financial product.

[0084] The implementation of steps S201-S204 is the same as that of the above steps S101-S104, and will not be described in detail here.

[0085] Step S205: Segment the news and public opinion data.

[0086] In one implementation, the server can sort the news and public opinion data in chronological order based on the time to which the news and public opinion data belongs, input the sorted news and public opinion data into a pre-trained large language model for summary extraction, and obtain news summary text of the news and public opinion data. Then, a preset prompt is used to instruct the large language model to segment the news summary text, thereby obtaining multiple paragraphs of news summary text as the multiple paragraphs of news and public opinion data. The preset prompt may include: segmenting the news summary text with approximately 100 words per paragraph, and the preset prompt may also include: segmenting the news summary text according to different topics. The length of the news summary text may also be limited to a range of 600 to 1000 words.

[0087] For example, if the obtained news summary text is 700 words long, and the news summary text is segmented with each paragraph having about 100 words, 7 paragraphs of news public opinion data can be obtained.

[0088] In another implementation method, the server can obtain the news summary text in the manner of the previous implementation method, then extract the semantics in the news summary text, and segment the news summary text according to preset semantic categories. For example, the preset semantic categories may include: social evaluation and public opinion, financial market news, policy news, etc.

[0089] In another implementation, the server may segment the obtained text of the news and public opinion data according to a preset paragraph word limit to obtain multiple paragraphs of news and public opinion data.

[0090] Step S206: for each paragraph of news and public opinion data obtained by segmentation, convert the news and public opinion data of the paragraph into text data in a preset data format.

[0091] In one implementation, when the preset data format is a string array, the server converts the obtained news and public opinion data of each paragraph into text data in the preset data format, and obtains text data in the string array data format corresponding to each paragraph.

[0092] For example, if news and public opinion data is segmented into 7 paragraphs, the server converts the data format of each paragraph of news and public opinion data to obtain a string array containing the text data of the 7 paragraphs, and the length of the string array is 7. The length of the string array is the same as the number of paragraphs obtained by segmenting the news and public opinion data.

[0093] Step S207: for each paragraph of text data obtained, generate an audio segment of the voice that announces the text data of the paragraph.

[0094] In one implementation, when the preset data format is a string array, for each paragraph of text data obtained, the server can generate an audio segment of the voice broadcasting the paragraph's text data in the following manner: the server can input the text data in the string array data format corresponding to the paragraph into a preset speech generation model to generate an audio segment of the voice broadcasting the paragraph's text data. The audio segments of each paragraph obtained can have a sequential order, for example, the audio segments of each paragraph can be sorted according to the order of the text data of each paragraph in the news and public opinion data.

[0095] For example, if a string array of 7 paragraphs is obtained, the string array of the first paragraph is input into the preset speech generation model to generate an audio segment of the speech that broadcasts the text data of the first paragraph, and so on, until an audio segment of the speech that broadcasts the text data of the seventh paragraph is obtained.

[0096] Step S208: For each paragraph of news public opinion data, generate a background image of the paragraph based on the text data of the paragraph.

[0097] In one implementation, for each paragraph of news and public opinion data, the server can generate a background image for that paragraph using the following method: The server can input the paragraph's text data into a modified stable-diffusion model (a generative model). The modified stable-diffusion model can then output a background image related to the content described in the paragraph's text data. For example, if the text data primarily describes a shareholder meeting at XX Company, the modified stable-diffusion model might output a background image of multiple people in a conference room. The modified stable-diffusion model can be fine-tuned using financial graphic data.

[0098] The following is an example of generating the background image for each paragraph.

[0099] For example, if step S207 obtains text data for seven paragraphs, the server can input the first paragraph into the stable-diffusion improved model to obtain the background image for the first paragraph, and so on, until the seventh paragraph is input into the stable-diffusion improved model to obtain the background image for the seventh paragraph, ultimately obtaining background images for seven paragraphs. The number of generated background images is the same as the number of paragraphs obtained by segmenting the news and public opinion data.

[0100] Furthermore, the server may generate a background image of a specified resolution according to a preset background resolution. For example, the background resolution may be 1280×720.

[0101] Step S209: Generate video frames of the background video layer of the target video based on the duration of the audio segment of each paragraph and the background image of each paragraph.

[0102] The server can generate video frames of the background video layer of the target video by following the steps AB:

[0103] Step A: For each paragraph, generate an entry and exit special effects video frame for the paragraph's background image. Between the entry and exit special effects video frames, generate a video frame displaying the paragraph's background image, thereby obtaining segmented video frames for the paragraph. The duration of the generated video frame displaying the paragraph's background image is the same as the duration of the paragraph's audio segment. The order of generating the entry and exit special effects video frames and generating the video frame displaying the paragraph's background image is not crucial.

[0104] In one implementation, for each paragraph, the server can generate segmented video frames of the paragraph according to the preset parameters of the background video layer video:

[0105] Generate a video frame showing the background image of the paragraph with a video duration equal to the duration of the audio segment of the paragraph, and in accordance with the video playback order, generate a background entry special effect video frame of the paragraph with the duration of the background entry special effect before the video frame of the above-mentioned background image, and generate a background entry special effect video frame of the paragraph with the duration of the background exit special effect after the video frame of the above-mentioned background image, thereby obtaining segmented video frames of the paragraph. The video parameters of the background video layer may include: background frame rate, background entry special effect duration, background exit special effect duration, etc. For example, the background frame rate may be 24 frames per second, and the background entry special effect duration and background exit special effect duration may be 0.5 seconds.

[0106] For another example, the server can use ffmpeg (an open source multimedia processing framework) to generate segmented video frames for each paragraph, such as by following the following instructions: ffmpeg -framerate 24 -i level1_imgs.txt -vf "transition=dissolve,0.5:0.5" -c:v libx264 -pix_fmt yuv420p -crf23 level1.mp4. Among them, -framerate 24 means setting the video frame rate to 24 frames per second, -framerate 24 means setting the background frame rate to 24 frames per second, -i level1_imgs.txt: indicates the text file that stores the file names of each background image and the absolute path of the file storage, -vf "transition=dissolve,0.5:0.5" indicates that the entry and exit effects are set as: dissolve effect to field transition effect, 0.5:0.5 means setting the background entry effect duration to 0.5 seconds and the background exit effect duration to 0.5 seconds, -c:v libx264 indicates using the libx264 encoder for video encoding, -pix_fmt yuv420p indicates setting the pixel format to yuv420p, -crf 23 indicates setting the output video quality factor to 23, and level1.mp4 indicates the name of the set output video file.

[0107] In this way, the background image in the video can be switched more smoothly and naturally, further improving the visual effect and viewing experience of the generated target video.

[0108] Step B: Splicing the segmented video frames of each generated paragraph to obtain the video frames of the background video layer of the target video.

[0109] In one implementation, the segmented video frames of each paragraph may be sorted according to the sorting order of the audio segments of each paragraph, and the generated segmented video frames of each paragraph may be spliced to obtain the video frames of the background video layer of the target video.

[0110] Step S210: Based on the total duration of each audio segment and the financial chart image, a video frame of the information video layer of the target video is generated to obtain the video frame of the target video.

[0111] The image of the information video layer is displayed on top of the image of the background video layer.

[0112] In one implementation method, the server can set the video frames of the background video layer of the target video and the video frames of the information video layer of the target video in different layers and merge them to obtain the video frames of the target video, wherein the layer where the video frames of the information video layer of the target video are located is below the layer where the video frames of the background video layer of the target video are located.

[0113] Step S211: Synchronize the audio track of the audio segment with the video frame of the target video to obtain the target video, and feed back the target video to the client.

[0114] When the video frames of the background video layer of the target video are obtained according to the above steps AB, the video duration of the video frames of the background image of each paragraph is the same as the duration of the audio segment of the paragraph. After adding the entry special effect video frames and the exit special effect video frames, the audio track synchronization can be performed in the following way: add interval audio of the audio interval time between the audio segments of two adjacent paragraphs.

[0115] The audio interval duration is the sum of the duration of the first segment's exit effect and the duration of the second segment's entrance effect. The first segment is the segment containing the video frames that play first in the video playback order, and the second segment is the segment containing the video frames that play later in the video playback order. The interval audio can be blank audio or the special effect audio for the entrance and exit effects.

[0116] For example, if you use the following command to generate segmented video frames for each paragraph: ffmpeg -framerate 24 -ilevel1_imgs.txt -vf "transition=dissolve,0.5:0.5" -c:v libx264 -pix_fmtyuv420p -crf 23 level1.mp4, the duration of the entry and exit effects in each paragraph's video frames is 0.5 seconds. That is, the exit effect in the first paragraph's video frames lasts 0.5 seconds, and the entry effect in the second paragraph's video frames lasts 0.5 seconds. The audio interval is 0.5 seconds + 0.5 seconds = 1 second.

[0117] In addition, if the entry effect of the segmented video frame of the first paragraph is 0.5 seconds, then the timestamp of the start of the audio segment of the first paragraph should be aligned with the timestamp of the start of the segmented video frame of the first paragraph. An opening video frame displaying text information such as the target financial product ID and other information of the target financial product can also be added before the segmented video frame of the first paragraph, and a video frame displaying risk warnings can be added at the end of the target video. In this way, by setting the interval audio time, the visual effect and viewing experience of the generated target video can be further improved, and the situation where the audio track is out of sync with the video screen can be reduced.

[0118] As can be seen from the above, by segmenting the news and public opinion data, audio segments and background images can be generated for different paragraphs. By generating video frames of the background video layer and the information video layer, the visual effect and viewing experience of the generated target video can be improved.

[0119] The following describes how the server obtains financial data and generates financial chart images in steps S102-S103.

[0120] In one embodiment of the present invention, the financial data may include at least one of the following data:

[0121] 1. Intraday financial data of the target financial product on the target trading day.

[0122] 2. Intraday financial data of the target financial product during the trading day interval, where the trading day interval is the first preset number of consecutive trading days including the target trading day.

[0123] 3. Financial data from financial reports of companies related to the target financial products.

[0124] In this way, by obtaining multi-dimensional financial data, users can be provided with richer financial information, increase their sources of financial information acquisition, and further enable users to efficiently obtain multi-dimensional financial information, thereby improving the efficiency of users in obtaining financial information.

[0125] The following describes how to obtain different financial data.

[0126] 1. Intraday financial data of the target financial product on the target trading day

[0127] The intraday financial data is obtained in the following manner: for each first preset time period in the target trading day, based on the trading data of the target financial product in the first preset time period, the first trading indicator of the target financial product in the first preset time period is obtained, and the intraday financial data including the first trading indicator of each first preset time period in the target trading day is obtained.

[0128] The first transaction indicator may include at least one of the following indicators: the starting price of the first preset time period, the ending price of the first preset time period, the highest price within the first preset time period, and the lowest price within the first preset time period. For example, the first preset time period may be one minute, five minutes, or ten minutes.

[0129] In one implementation, for each first preset time period on a target trading day, the server may determine, based on the timestamp of the transaction data, the transaction data within the first preset time period, and calculate the first transaction index of the target financial product for the first preset time period based on the transaction data within the first preset time period. For example, taking the first preset time period as one minute, the server may determine the transaction data within [10:40:00, 10:41:00), and based on the transaction data within [10:40:00, 10:41:00), calculate the first transaction index of the target financial product for [10:40:00, 10:41:00]. Similarly, calculate the first transaction index of the target financial product for [10:41:00, 10:42:00), and so on, until the first transaction index of the target financial product for each minute of the trading time period on the target trading day is calculated, thereby obtaining the first transaction index at the minute level.

[0130] The above step S103 can generate a financial chart image of intraday financial data in the following manner:

[0131] For each first trading indicator of a first preset time period in a target trading day, a financial chart image including the first trading indicator of the first preset time period and the first trading indicator of the preceding time period of the first preset time period is generated according to the time sequence of each first preset time period in the target trading day.

[0132] Specifically, the server generates a financial chart image of the intraday financial data of each first preset time period for each first trading indicator of the first preset time period. Then, according to the time sequence of each first preset time period in the target trading day, the server obtains the financial chart image of the intraday financial data of the first preset time period. When generating the first trading indicator of the second first preset time period, the server adds the first trading indicator of the second first preset time period to the first trading indicator of the first preset time period used for the financial chart image of the intraday financial data of the first preset time period, and generates a financial chart image including multiple consecutive first trading indicators of the first preset time period according to the added first trading indicator, thereby obtaining the financial chart image of the second intraday financial data. Similarly, when generating the first trading indicator of the current first preset time period, the server adds the first trading indicator of the current first preset time period to the first trading indicator of the first preset time period used for the financial chart image of the intraday financial data of the previous first preset time period, and generates a financial chart image including multiple consecutive first trading indicators of the first preset time period according to the added first trading indicator, until a financial chart image of the kth intraday financial data is generated, where k represents the number of first preset time periods included in the target trading day. In the process of generating the financial chart image, the image may be generated according to pre-set image parameters of the financial chart image. For example, the image parameters may include resolution, and the resolution of the financial chart image may be set to 640×360.

[0133] For example, see Figure 3a Schematic diagram of the first type of financial chart image provided, Figure 3b A diagram of the second type of financial chart image provided, Figure 3c The third type of financial chart image provided is shown in the figure below. Figure 3d The fourth type of financial chart image provided is shown, assuming Figure 3a is the i-th financial chart image, then the i+1-th financial chart image can be found in Figure 3b The financial chart image shown, the i+2th financial chart image can be found in Figure 3c The financial chart image shown, the i+3th financial chart image can be found in Figure 3d Financial chart image shown.

[0134] The above step S103 may also generate a financial chart image of intraday financial data in the following manner:

[0135] After obtaining the k financial chart images of intraday financial data according to the above implementation, the server can draw a dynamic picture based on the k financial chart images of intraday financial data as the financial chart image of intraday financial data.

[0136] Specifically, the server can use FFmpeg to draw dynamic images based on financial chart images of k intraday financial data.

[0137] For example, the server can sequentially name the generated financial chart images for k intraday financial data: image1.jpg, image2.jpg, and imagek.jpg. The file names and absolute paths of the generated image files for the k intraday financial data chart images are recorded in a text file named images.txt. Then, a dynamic image can be drawn based on the k intraday financial data chart images using the following command: ffmpeg -f image2pipe -i images.txt -r 24 -loop 1 output.gif. Here, -fimage2pipe instructs ffmpeg to read the input file in image pipe format, -r 24 sets the frame rate of the dynamic image to 24 frames per second, -loop 1 indicates that the dynamic image is played in a loop, and output.gif is the file name of the generated dynamic image. Furthermore, if you need to adjust the size or color depth of the dynamic image, you can add additional ffmpeg parameters, such as -s (to set the resolution) and -pix_fmt (to set the pixel format).

[0138] 2. Intraday financial data of the target financial product during the trading day

[0139] The intraday financial data is obtained in the following manner: for each trading day in the trading day interval, based on the trading data of the target financial product on that trading day, the second trading indicator of the target financial product on that trading day is obtained, and the intraday financial data including the second trading indicator of each trading day in the trading day interval is obtained.

[0140] The second trading indicator may include at least one of the following indicators: the highest price of the trading day, the lowest price of the trading day, the opening price of the trading day, the closing price of the trading day, and the trading volume of the trading day. The trading day interval may be 22 trading days. If the target trading day is the trading day closest to the current date, the trading day interval may be the target trading day and the 21 trading days preceding the target trading day.

[0141] In one implementation, for each target trading day, the server can determine the trading data belonging to that trading day based on the timestamp of the trading data, and calculate the second trading indicator of the target financial product for that trading day based on the trading data. Furthermore, if the target financial product is a stock, and the stock undergoes ex-rights or ex-dividend changes, the server can use the ex-rights adjustment, unaffordable, and ex-rights adjustment methods to calculate the ex-rights adjustment, unaffordable, and ex-rights adjustment second trading indicators for the trading day interval, respectively. A corresponding financial chart image can also be generated for each second trading indicator.

[0142] The above step S103 can generate a financial chart image of the intraday financial data in the following manner:

[0143] For the second trading indicator of each trading day in the target trading day, a financial chart image including the second trading indicator of the trading day and the second trading indicator of the previous time of the trading day is generated according to the time sequence of each trading day in the trading day interval.

[0144] Specifically, the server generates a financial chart image of the intraday financial data of each trading day for the second trading indicator of the trading day, and then obtains the financial chart image of the intraday financial data of the first trading day according to the chronological order of each trading day in the trading day interval. When generating the second trading indicator of the second trading day, the server adds the second trading indicator of the second trading day based on the second trading indicator of the trading day used in the financial chart image of the intraday financial data of the first trading day, and generates a financial chart image including the second trading indicators of multiple consecutive trading days according to the added second trading indicator to obtain the financial chart image of the second intraday financial data. And so on, when generating the second trading indicator of the current trading day, the server adds the second trading indicator of the trading day used in the financial chart image of the intraday financial data of the previous trading day based on the second trading indicator of the trading day, and generates a financial chart image including the second trading indicators of multiple consecutive trading days according to the added second trading indicator, until the financial chart image of the j-th intraday financial data is generated, where j represents the number of trading days included in the trading day interval. In the process of generating the financial chart image, the image can be generated according to the preset image parameters of the financial chart image. For example, the image parameters may include resolution. For example, the resolution of the financial chart image can be set to 640×360. In addition, the generated financial chart image of the intraday financial data can also refer to the following: Figure 3a-3d Financial chart image shown.

[0145] The above step S103 may also generate a financial chart image of the intraday financial data in the following manner:

[0146] After obtaining the j financial chart images of the daily financial data according to the above implementation, the server can draw a dynamic picture based on the j financial chart images of the daily financial data as the financial chart image of the daily financial data.

[0147] Specifically, the server may use FFmpeg to draw dynamic images based on financial chart images of j daily financial data.

[0148] For example, the server can sequentially name the generated financial chart images for each of the j daily financial data periods: image1.jpg, image2.jpg, and imagej.jpg. The server can also record the file names and absolute paths of the generated images in a text file named images.txt. The server can then render a dynamic image based on the generated images using the following command: ffmpeg -f image2pipe -i images.txt -r 24 -loop 1 output.gif.

[0149] 3. Financial data of companies related to the target financial products

[0150] The financial report financial data is obtained in the following manner: for each financial reporting period within a first historical time period of a company, based on the company's financial report during that reporting period, the company's financial report indicators for that reporting period are obtained, thereby obtaining financial report financial data including the company's financial report indicators for each reporting period. The first historical time period may be a preset length of time prior to a target trading date. For example, the preset length of time may be three years, in which case the first historical time period is three years prior to the target trading date. Each financial reporting period may be annual, semi-annual, or quarterly. The company's financial reports for that reporting period used to calculate the financial report indicators may include annual, semi-annual, and quarterly reports.

[0151] In one implementation, the server may calculate at least one of the following indicators for the financial reporting period as the company's financial reporting indicator for the financial reporting period: total revenue amount, net profit attributable to parent company amount, debt-to-net asset ratio, equity ratio, and tangible net worth debt ratio.

[0152] The debt-to-equity ratio is obtained using the following expression: d2e = tot_liab / (tot_assets - tot_liab), where tot_liab is the company's total liabilities reported for the reporting period, and tot_assets is the company's total assets reported for the reporting period.

[0153] The equity ratio is obtained as follows: debttoequity = tot_liab / tot_shrhldr_eqy_excl_min_int, where tot_shrhldr_eqy_excl_min_int is the company's reported shareholders' equity for the reporting period.

[0154] The debt-to-tangible net worth ratio is obtained as follows:

[0155] ttdtotangibleequity_1 = tot_liab / (tot_shrhldr_eqy_excl_min_int - intang_assets), where tot_shrhldr_eqy_excl_min_int is the company's reported shareholders' equity for the reporting period, and intang_assets is the company's reported intangible assets for the reporting period.

[0156] After calculating the above-mentioned debt-to-net-asset ratio, property rights ratio, and tangible net worth debt ratio, each of the debt-to-net-asset ratio, property rights ratio, and tangible net worth debt ratio can be constructed into structured data in the form of a data frame in the form of quarterly, semi-annual, and annual data.

[0157] The above step S103 may generate a financial chart image of the financial report financial data in the following manner: for each financial report indicator of the company within the first historical period, generate a financial chart image representing the financial report indicator within the first historical period.

[0158] Specifically, the server can generate a bar chart of annual total revenue in the past three years, a bar chart of annual net profit attributable to shareholders in the past three years, a line chart of quarterly total revenue in the past three years, a line chart of quarterly net profit attributable to shareholders in the past three years, a line chart of changes in the debt-to-net asset ratio in each quarter in the past three years, a line chart of changes in the property rights ratio in each quarter in the past three years, and a line chart of changes in the tangible net worth debt ratio in the past three years.

[0159] The above step S103 can also generate a financial chart image of the financial report financial data in the following manner: for each financial report indicator of the latest financial report period within the company's first historical time, generate a year-on-year financial chart image representing the financial report indicator of the latest financial report period relative to the financial report indicator of the same financial report period in the previous year.

[0160] Specifically, the server can generate a year-on-year bar chart of quarterly total revenue for the quarter in which the most recently released financial report is located, and a year-on-year bar chart of quarterly net profit attributable to shareholders for the quarter in which the most recently released financial report is located.

[0161] In addition, the server can generate financial chart images of financial report financial data based on matplot or highcarts (two visual chart drawing tools).

[0162] When generating a financial chart image of financial data in a financial report, the resolution of the financial chart image can be set to 640×360.

[0163] In addition, before obtaining financial data, initial data for each type of financial data can be obtained. For example, the initial data for intraday financial data and interday financial data can include the transaction data of the target financial product, and the initial data for financial report financial data can include the company's financial report. The data source and data storage format of the initial data of each type of financial data are different. The server can first obtain the initial data of the above three types of financial data respectively according to different preset financial data acquisition interfaces, and then classify and identify the initial data of each type of financial data obtained. Then, based on the classification identification, determine the method for processing the initial data of the financial data, and then process the initial data to obtain the financial data. In this way, the initial data including structured information and unstructured information can be processed accordingly to obtain financial data, which facilitates further pre-processing of the financial data such as cleaning, sorting, and analysis.

[0164] In this way, by pre-processing the obtained multi-dimensional financial data such as cleaning, sorting and analysis, the financial information can be further refined, and more intuitive and detailed financial information can be provided to users, further enabling users to efficiently obtain multi-dimensional financial information and improve the efficiency of users in obtaining financial information.

[0165] The following describes the method of generating the video frame of the target video in step S107.

[0166] In one embodiment of the present invention, see Figure 4 A flowchart of a method for generating a video frame is provided, wherein the method includes the following steps S401-S403.

[0167] Step S401: Obtaining the information level of the material used to generate the video frame of the target video.

[0168] The material includes at least one of the following information: a financial chart image, text data, and an image generated based on the text data, and the information level represents the amount of financial information contained in the material.

[0169] In one implementation, a material information content level is pre-set for each type of material, and the information content level value of each type of material can be recorded. For example, the information content level of text data can be 1, the information content level of an image generated based on the text data can be 2, and the information content level of a financial chart image of intraday financial data can be 3, and the information content level can be 4. The higher the value of the material information content level, the more financial information the material contains. The server can calculate the information content level value of each type of material based on the recorded information content level value.

[0170] For example, assuming the materials include: a financial chart image generated based on a company's financial report, a financial chart image generated based on intraday financial data, a financial chart image generated based on daytime financial data, and an image generated based on text data, then the information content level of the financial chart image generated based on the company's financial report is level three, the information content level of the financial chart image generated based on intraday financial data is level two, the information content level of the financial chart image generated based on daytime financial data is level two, and the information content level of the image generated based on text data is level one. In descending order of information content, the information content levels are level one, level two, and level three.

[0171] In another implementation, the server can determine the time span of the data used to generate the material, and sort the material in descending order based on the time span of the data used to generate the material, thereby obtaining the information content level of the material used to generate the video frames of the target video. For example, for the material of financial chart images of intraday financial data, the material used to generate the financial chart images of the intraday financial data is the intraday financial data of the first trading indicator for each first preset time period on the target trading day, and the time span is the target trading day. For the material of financial chart images generated based on financial report financial data, the material used to generate the financial chart images is the financial report of the company for each financial reporting period within a first historical time period, and the time span is the first historical time period, which is one day longer than the target trading day. Therefore, the information content level of the material of financial chart images generated based on financial report financial data can be greater than that of the material of financial chart images of intraday financial data.

[0172] If the materials include: financial chart images generated based on a company's financial report, financial chart images generated based on intraday financial data, and images generated based on text data, then sorting the materials in descending order of time span yields the following order: financial chart images generated based on a company's financial report, financial chart images generated based on intraday financial data, and images generated based on text data. According to this order, the information content level of the financial chart images generated based on the company's financial report is ranked third, the information content level of the financial chart images generated based on intraday financial data is ranked second, and the information content level of the images generated based on text data is ranked first.

[0173] Other implementations of step S401 for obtaining the information level of the material will be described in the following embodiments and will not be described in detail here.

[0174] Step S402: Based on the information level of each material, determine the layer of each material in the video frame and the layout information in the layer.

[0175] In one implementation method, a preset correspondence between the material belonging to each information level and the layers in the video frame and the layout information in the layers can be set for each information level. The server can use the above preset correspondence to quickly generate video frames of the target video using relatively fixed materials in the layers in the video frame and the layout information in the layers, thereby improving the efficiency of video generation and the efficiency of pushing videos to users.

[0176] For example, the layer and layout information of the material in the video frame corresponding to the material with the first information level are: located in the bottom layer. The layer and layout information of the material in the video frame corresponding to the material with the second information level are: located in the layer above the bottom layer, and located in the preset low-attention area of the layer. The layer and layout information of the material in the video frame corresponding to the material with the second information level are: located in the layer above the bottom layer, and located in the preset high-attention area of the layer. For example, the low-attention area can include an area of a preset size in the upper left corner and an area of a preset size in the upper right corner, and the high-attention area can include an area of a preset size in the middle. For example, the low-attention area can include an area located in the upper left corner that is 1 / 4 the size of the video screen and an area located in the upper right corner that is 1 / 4 the size of the video screen.

[0177] For another example, the display area of the material corresponding to the third level of information level in the layer is: located in the central area of the target video, with the center of the target video as the center of the circle and the length or width of the target video at the first preset ratio as the diameter. The display area of the material corresponding to the second level of information level in the layer is: located in the four corners of the target video, with the four corners of the target video as the center of the circle and the length or width of the target video at the second preset ratio as the radius. The display area of the material corresponding to the first level of information level in the layer is: located in the edge area of the target video, with the four edges of the target video as the starting ends and the area formed by the length or width of the target video at the third preset ratio extending to the center of the target video. For example, the first preset ratio can be 0.35, the second preset ratio can be 0.15, and the third preset ratio can be 0.10.

[0178] Furthermore, for materials with a third-level information content rating, you can set prominent effects such as dynamic magnification and three-dimensional floating effects. For example, the zoom factor for dynamic magnification can be set to 1.2-1.5. For materials with a second-level information content rating, you can use dynamic thumbnails for display. For materials with a first-level information content rating, you can use a semi-transparent mode for carousel display, for example, the transparency can be set to 60%.

[0179] Step S403: Generate a video frame of the target video based on the layer where each material is located, the layout information of the material in the layer, and the duration of the audio segment.

[0180] In one implementation, for materials with different information content levels, the server can determine the layer and layout information corresponding to the information content level of the material, and use the material in the layout area indicated by the layout information in the determined layer to generate video frames corresponding to the information content level. This process continues until video frames corresponding to each information content level are generated, resulting in the video frames of the target video. The video duration of the generated video frames is the same as the duration of the audio segments. In addition, if there are multiple audio segments, the video duration of the generated video frames is the same as the total duration of each audio segment.

[0181] For example, the materials include: financial chart images generated based on the company's financial report, financial chart images generated based on intraday financial data, financial chart images generated based on interday financial data, and images generated based on text data. Moreover, the information volume level of the financial chart images generated based on the company's financial report is the third level, the information volume level of the financial chart images generated based on intraday financial data is the second level, the information volume level of the financial chart images generated based on interday financial data is the second level, and the information volume level of the images generated based on text data is the first level.

[0182] In the process of generating video frames using materials of different information levels, for each information level, video frames can be generated based on the materials of that information level and in accordance with the pre-set video parameters of that information level. For example, if the video parameters include the video frame rate, entry effect duration, exit effect duration, and transition effects, for video frames generated using images generated based on text data with a first information level, the frame rate of the generated video frames can be set to 24 frames per second, and the entry effect duration and exit effect duration can both be set to 0.5 seconds. For video frames generated using financial chart images generated based on intraday financial data and financial chart images generated based on daytime financial data with a second information level, the frame rate of the generated video frames can be set to 48 frames per second, and the entry effect duration and exit effect duration can both be set to 1 second. For video frames generated using financial chart images generated based on a company's financial report with an information level of the third level, the frame rate of the generated video frames can be set to 60 frames per second, and the duration of the entry and exit special effects are both 1.5 seconds.

[0183] The instructions for generating video frames of different information levels based on materials of different information levels can refer to the above step S209. For example, the server can generate video frames of the first information level according to the following instructions: ffmpeg -framerate 24 -i level1_imgs.txt -vf "transition=dissolve,0.5:0.5" -c:vlibx264 -pix_fmt yuv420p -crf 23 level1.mp4. For example, -i level1_imgs.txt can represent a text file that stores the file names and absolute paths of materials of the first information level.

[0184] The server can generate video frames with a Level 2 information level using the following command: ffmpeg -framerate 48 -i level2_imgs.txt -vf "transition=rectcrop,1:1" -c:v libx264 -pix_fmt yuv420p -crf 23 level2.mp4. For example, "-i level2_imgs.txt" can represent the file name and absolute path of each Level 2 source file. "-vf "transition=rectcrop,1:1" specifies a rectangular cropping transition effect for the entry and exit effects, and "1:1" specifies a 1-second entry and 1-second exit duration for the effect.

[0185] The server can generate video frames with a Level 3 information content level using the following command: ffmpeg -framerate 60 -i level3_imgs.txt -vf "transition=slideright,1.5:1.5" -c:vlibx264 -pix_fmt yuv420p -crf 23 level2.mp4. For example, "-i level2_imgs.txt" can represent the file name and absolute path of each Level 3 source file. "-vf " transition=rectcrop,1:1" specifies a right-sliding transition effect for the entry and exit effects. "1.5:1.5" specifies a 1.5-second entry and 1.5-second exit duration for the entry and exit effects, respectively.

[0186] In this way, according to the information level of the material, the video frames generated using materials with different information levels are divided into information levels. When generating the video frames of the target video, the parameters can be flexibly adjusted according to the information level to highlight more important information, thereby enabling users to more efficiently obtain more information with higher information volume and more important information, thereby enhancing the visual effect and viewing experience.

[0187] The following describes the method of obtaining the information level of the material in step S401.

[0188] In one embodiment of the present application, the server can obtain the information volume representation value of the material based on the financial data or news and public opinion data used to generate the material, and then determine the information volume level of the material based on the information volume representation value of the material.

[0189] The following describes how to determine the information representation value of a material in different situations.

[0190] Case 1: The data used to generate the material is news and public opinion data

[0191] In one implementation, the server can determine the timeliness importance of the news and public opinion data based on the data's release date, a preset timeliness enhancement coefficient, and a preset time decay coefficient. It then obtains the information content representation of the material based on the frequency representation of each financial keyword in the news and public opinion data, the amount of text in the news and public opinion data, and the timeliness representation.

[0192] The financial keywords in news and public opinion data are those obtained by keyword matching the news and public opinion data based on the financial keywords in a pre-set financial entity vocabulary. The term frequency representation of the financial keywords is a numerical value that represents the frequency of the financial keywords in the news and public opinion data and their representativeness to the news and public opinion data. For example, the term frequency representation can be determined using the TF-IDF (term frequency–inverse document frequency) method. The timeliness enhancement coefficient is a coefficient set based on the degree to which time affects the importance of the news and public opinion data. For example, a higher timeliness enhancement coefficient can be set for news and public opinion data with strong timeliness. The time decay coefficient is a coefficient set based on the degree to which the importance of news and public opinion data decays over time. For example, if the time since the release of news and public opinion data increases, the importance of the news and public opinion data decreases, that is, the reference value of the news and public opinion data decreases. In this case, a larger time decay coefficient can be set.

[0193] Specifically, the server can determine the information content representation value of the material according to the following expression:

[0194] ;

[0195] in, Financial keywords in the news and public opinion data belong to the preset financial entity vocabulary. Represents financial keywords determined based on the TF-IDF method The word frequency representation value of is the preset time-effectiveness reinforcement coefficient, is the preset time attenuation coefficient, The time interval between the release time of news and public opinion data and the time when the user sends the video request. This is the timeliness representation value. In this way, the timeliness attenuation of news and public opinion data can be achieved. When the time between the release of news and public opinion data and the time when the user sends the video request is too long, the information representation value of the news and public opinion data is reduced, thereby reducing the importance of the material generated based on the news and public opinion data in the target video. For example, if For a hour, the information representation value of the material calculated is Q, then if If it is a+24 hours, the information representation value of the material calculated is 0.1Q.

[0196] Case 2: The data used to generate the material is the financial data of the company related to the target financial product

[0197] In one implementation, the server can determine, for each financial indicator in the financial report data, the information value corresponding to the financial indicator based on the difference between the current value and the base value of the financial indicator, the dispersion value of the financial indicator in the preset historical financial reporting period, the weight corresponding to the financial indicator, and the industry deviation of the industry to which the relevant company of the target financial product belongs.

[0198] Among them, the financial reporting indicators of the financial reporting data may include at least one of the following indicators: total revenue amount, net profit attributable to shareholders, debt-to-equity ratio, property rights ratio and tangible net worth debt ratio. The current value of the financial reporting indicator is: the value of the financial reporting indicator in the latest financial reporting period. The base period value of the financial reporting indicator is: the value of the financial reporting indicator in the preset benchmark historical financial reporting period. The preset historical financial reporting period can be: the second number of consecutive financial reporting periods in history including the financial reporting period to which the target trading day belongs. For example, the second number can be 12, and the dispersion representation value can be the standard deviation.

[0199] After obtaining the information quantity representation value corresponding to each financial report indicator, the sum of the information quantity representation values corresponding to each financial report indicator can be used as the information quantity representation value of the material.

[0200] For example, the server can determine the information content representation value of the material according to the following expression:

[0201] ;

[0202] Where n is the number of types of financial indicators in financial report data, is the difference between the current value and the base period value of the i-th financial report indicator, is the dispersion representation value of the i-th financial report indicator in the preset historical financial report period, is the weight corresponding to the i-th financial report indicator, for The sigmoid (S-shaped growth curve) function, The preset industry deviation.

[0203] For example, the value of the sigmoid function can be calculated as follows: , is the steepness coefficient, such as, It can be 5. The weight corresponding to the i-th financial report indicator can be set according to the type of the i-th financial report indicator. For example, if the i-th financial report indicator belongs to the asset and liability type, It can be set to 1.2. If the i-th financial report indicator belongs to the profit type, Can be set to 1, if the i-th financial indicator belongs to the cash flow type, Can be set to 0.8.

[0204] Case 3: The data used to generate the material is the intraday financial data and / or interday financial data of the target financial product

[0205] In one implementation, the server can obtain the information content representation value of the material based on the maximum transaction volume within each time unit of the quantity-price financial data, the number of times the quantity-price financial data meets a preset quantity-price change trend, and the transaction volume and rate of return within each time unit of the quantity-price financial data. The quantity-price financial data may be intraday financial data or interday financial data. If the quantity-price financial data is intraday financial data, the time unit may be a first preset duration; if the quantity-price financial data is interday financial data, the time unit may be a single trading day. The preset quantity-price change trend may be a breakout pattern change trend that meets the golden cross or dead cross conditions of the MACD (Moving Average Convergence Divergence).

[0206] In addition, when the data used to generate the material is the intraday financial data and interday financial data of the target financial product, the information quantity representation value of the material of the intraday financial data can be determined for the material generated based on the intraday financial data of the target financial product, and the information quantity representation value of the material of the interday financial data can be determined for the material generated based on the interday financial data of the target financial product.

[0207] Specifically, the server can determine the information content representation value of the material according to the following expression:

[0208] ;

[0209] in, is the number of time units included in the quantity and price financial data, is the trading volume of the t-th time unit, is the rate of return of the t-th time unit, It is the maximum value of the transaction volume in each time unit in the quantity and price financial data. It is the number of times that the quantity and price financial data meets the preset quantity and price change trend.

[0210] In the expression, use To calculate the information representation value of the material instead of using pure yield, reduce the impact of low trading volume on the accuracy of determining the number of times that the preset volume and price change trend is met.

[0211] The following describes a method for determining the information level of a material based on the information representation value of the material.

[0212] In one implementation, the server may determine the information volume score of each material according to the information volume representation value of each material, and determine the information volume level of the material based on the correspondence between the information volume score of the material, a preset information volume score and the information volume level.

[0213] The server can determine the information content score of each material based on the maximum and minimum values of the information content representation value of each material. For example, the server can determine the information content score of each material according to the following expression:

[0214] ;

[0215] in, For the The information content score of each material, For the The information value representation of a material, is the minimum value of the information representation value of each material, It is the maximum value of the information representation value of each material.

[0216] Assume that the correspondence between the preset information score and information level is: , then The information level of a material is the first level. , then The information level of the material is the second level. , then The information level of each material is the third level. and is the preset grade cutoff value, for example, It can be 0.5, It can be 0.8.

[0217] The following describes a method for determining the layout information of each material in the target video.

[0218] For each material, the server can determine the material's layout information in the video frame based on the material's information score, the target video's preset resolution, the preset horizontal distribution factor, and the preset vertical amplitude coefficient. For example, the coordinates of the material in the video frame can be determined as layout information in the following manner:

[0219] ;

[0220] ;

[0221] in, For the The horizontal coordinate of the material in the video frame, The width of the preset resolution of the target video. is the preset horizontal distribution factor, is the inverse error function, For the The vertical coordinate of the material in the video frame, The height of the preset resolution of the target video. is the preset vertical amplitude coefficient, For the The sequence number of the material in the same frame. The total number of materials in the same frame.

[0222] Determining the layout information in the above manner and then generating the video frames of the target video does not simply use fixed layout information to generate the video frames of the target video. Determining the layout information in the above manner can flexibly determine the layout information of each material according to the importance of different financial data in different time periods, for different target trading days and different target financial products, so that in the final generated target video, materials including financial data with higher importance will be highlighted and materials including financial data with lower importance will be weakened, thereby improving the user's video viewing experience and allowing the user to intuitively obtain more important financial information.

[0223] Furthermore, the server can determine a visual effect color compensation value for the material based on the material's color information, the background color, and the material's information content representation value. Based on the material's visual effect color compensation value, the primary color value of the material is adjusted, so that the color of the financial data material with a higher color compensation value is more prominent relative to the background color, and the color of the financial data material with a lower color compensation value is displayed less prominently relative to the background color.

[0224] Specifically, the server can calculate the visual effect color compensation value according to the following expression:

[0225] ;

[0226] in, For the The visual effect color compensation value of each material, For the The information value representation of a material, For the The contrast of the material, The maximum value of the contrast of each material.

[0227] It can be calculated as follows:

[0228] ;

[0229] in, is the brightness value of the main color of the material, is the brightness value of the background color, The red and green axis values of the main color of the material, Get the brightness value of the background color. Take the value of the yellow-blue axis of the main color of the material. Get the yellow-blue axis value for the background color.

[0230] By adjusting the main color value of the material in this way, the visual effect can make the color of the material including financial data with higher importance more prominent relative to the background color, highlighting the more important financial information, and the color of the material including financial data with lower importance is weakened relative to the background color, thereby improving the user's experience of watching the video.

[0231] Corresponding to the above-mentioned video push method applied to the server, the embodiment of the present invention also provides a video push method applied to the client. Figure 5 A flowchart of another video push method is provided, wherein the method includes the following steps S501-S503:

[0232] Step S501: Obtain the identifier of the target financial product and the target trading date input by the user.

[0233] For step S501 , reference may be made to the description of step S101 and will not be described in detail here.

[0234] Step S502: Send the identifier of the target financial product and the target trading day to the server, so that the server generates a target video according to the above-mentioned video push method applied to the server, and feeds back the target video to the client.

[0235] Step S503: Receive the target video and display the target video.

[0236] By applying the video push method provided by the embodiment of the present invention, when a user wants to obtain the financial information of a target financial product on a target trading day, the user can input the identifier of the target financial product and the target trading day on the user side and send a request to the server side. The server side collects the financial data of the target financial product and the news and public opinion data, processes the financial data to generate financial chart information, generates audio of the voice that broadcasts the Chinese characters in the news and public opinion data, and generates a target video that can introduce the financial information of the target financial product based on the audio and financial chart information, and feeds the generated target video back to the user side. By displaying the processed financial information to the user in the form of a dynamic and visual video, the user can obtain the target video that introduces the financial information of the target financial product in a more convenient way, which enables the user to obtain the financial information of the target financial product more quickly and easily, thereby improving the efficiency of the user in obtaining financial information.

[0237] In the technical solutions provided in the embodiments of the present application, operations such as acquisition, storage, use, processing, transmission, provision and disclosure of user personal information are all carried out with the user's authorization.

[0238] Corresponding to the above-mentioned video pushing method applied to the server, an embodiment of the present invention further provides a video pushing device.

[0239] In one embodiment of the present invention, see Figure 6 A schematic diagram of the structure of a video push device is provided, which is applied to a server and includes:

[0240] The first target information obtaining module 601 is used to obtain the identifier of the target financial product and the target trading date input by the user on the client.

[0241] The financial data acquisition module 602 is used to obtain the financial data of the target financial product based on the identifier and the target trading day.

[0242] The financial chart image generation module 603 is configured to generate a financial chart image of a target financial product based on the financial data.

[0243] The news and public opinion data acquisition module 604 is used to obtain news and public opinion data related to the target financial product.

[0244] The text data conversion module 605 is used to convert news and public opinion data into text data in a preset data format.

[0245] The audio segment generating module 606 is used to generate an audio segment of the voice that announces the text data.

[0246] The video frame generation module 607 is configured to generate video frames of a target video based on the duration of the audio segment and the financial chart image.

[0247] The target video feedback module 608 is used to synchronize the audio segments with the video frames of the target video to obtain the target video and feed back the target video to the client.

[0248] By applying the video push method provided by the embodiment of the present invention, when a user wants to obtain the financial information of a target financial product on a target trading day, the user can input the identifier of the target financial product and the target trading day on the user side and send a request to the server side. The server side collects the financial data of the target financial product and the news and public opinion data, processes the financial data to generate financial chart information, generates audio of the voice that broadcasts the Chinese characters in the news and public opinion data, and generates a target video that can introduce the financial information of the target financial product based on the audio and financial chart information, and feeds the generated target video back to the user side. By displaying the processed financial information to the user in the form of a dynamic and visual video, the user can obtain the target video that introduces the financial information of the target financial product in a more convenient way, which enables the user to obtain the financial information of the target financial product more quickly and easily, thereby improving the efficiency of the user in obtaining financial information.

[0249] In one embodiment of the present invention, the apparatus further comprises:

[0250] The segmentation processing module is used to process news and public opinion data in segments.

[0251] The text data conversion module is specifically used to convert the news and public opinion data of each paragraph obtained by segmentation into text data in a preset data format.

[0252] The audio segment generating module is specifically used to generate an audio segment of the voice that broadcasts the text data of each paragraph for the obtained text data.

[0253] The video frame generation module is specifically used to generate a background image of each paragraph based on the text data of the paragraph for the news and public opinion data of the paragraph; generate video frames of the background video layer of the target video based on the duration of the audio segment of each paragraph and the background image of each paragraph; generate video frames of the information video layer of the target video based on the total duration of each audio segment and the financial chart image, and obtain video frames of the target video, wherein the picture of the information video layer is displayed on the picture of the background video layer.

[0254] As can be seen from the above, by segmenting the news and public opinion data, audio segments and background images can be generated for different paragraphs. By generating video frames of the background video layer and the information video layer, the visual effect and viewing experience of the generated target video can be improved.

[0255] In one embodiment of the present invention, the video frame generation module is specifically used to generate, for each paragraph, an entry special effect video frame and an exit special effect video frame of the background image of the paragraph, and generate a video frame showing the background image of the paragraph between the entry special effect video frame and the exit special effect video frame to obtain segmented video frames of the paragraph, wherein the video duration of the generated video frame showing the background image of the paragraph is the same as the duration of the audio segment of the paragraph; the generated segmented video frames of each paragraph are spliced to obtain video frames of the background video layer of the target video.

[0256] In this way, the background image in the video can be switched more smoothly and naturally, further improving the visual effect and viewing experience of the generated target video.

[0257] In one embodiment of the present invention, the target video feedback module is specifically used to add interval audio of an audio interval time between the audio segments of two adjacent paragraphs, wherein the audio interval time is equal to the sum of the exit special effect duration of the first paragraph and the entrance special effect duration of the second paragraph. The first paragraph is: the paragraph to which the segmented video frame played first in the video playback order of the two adjacent paragraphs belongs, and the second paragraph is: the paragraph to which the segmented video frame played later in the video playback order of the two adjacent paragraphs belongs.

[0258] In this way, by setting the interval audio of the audio interval time, the visual effect and viewing experience of the generated target video can be further improved, and the situation where the audio track is out of sync with the video picture can be reduced.

[0259] In one embodiment of the present invention, a video frame generation module is specifically used to obtain the information content level of the material used to generate the video frame of the target video, wherein the material includes at least one of the following information: financial chart images, text data, and images generated based on the text data, and the information content level represents the amount of financial information contained in the material; based on the information content level of each material, the layer of each material in the video frame and the layout information in the layer are determined; based on the layer where each material is located, the layout information of the material in the layer, and the duration of the audio segment, the video frame of the target video is generated.

[0260] In this way, according to the information level of the material, the video frames generated using materials with different information levels are divided into information levels. When generating the video frames of the target video, the parameters can be flexibly adjusted according to the information level to highlight more important information, thereby enabling users to more efficiently obtain more information with higher information volume and more important information, thereby enhancing the visual effect and viewing experience.

[0261] In one embodiment of the present invention, the financial data includes at least one of the following data: intraday financial data of the target financial product on the target trading day, interday financial data of the target financial product in the trading day interval, and financial report financial data of companies related to the target financial product, wherein the trading day interval is: the first preset number of consecutive trading days including the target trading day.

[0262] In this way, by obtaining multi-dimensional financial data, users can be provided with richer financial information, increase their sources of financial information acquisition, and further enable users to efficiently obtain multi-dimensional financial information, thereby improving the efficiency of users in obtaining financial information.

[0263] In one embodiment of the present invention, when the financial data includes intraday financial data, the intraday financial data is obtained in the following manner: for each first preset time period in the target trading day, based on the trading data of the target financial product in the first preset time period, a first trading indicator of the target financial product in the first preset time period is obtained, and the intraday financial data including the first trading indicator for each first preset time period in the target trading day is obtained; based on the financial data, a financial chart image of the target financial product is generated, including: for each first preset time period in the target trading day, according to the time sequence of each first preset time period in the target trading day, a financial chart image including the first trading indicator for the first preset time period and the first trading indicator of the preceding time of the first preset time period is generated; and / or when the financial data includes inter-day financial data, the inter-day financial data is obtained in the following manner: for each trading day in the trading day interval, based on the trading data of the target financial product in the trading day, a second trading indicator of the target financial product on the trading day is obtained, and the intra-day financial data including the second trading indicator of each trading day in the trading day interval is obtained. generating a financial chart image of a target financial product based on the financial data, including: for each trading day within the trading day interval, generating, in chronological order within the trading day interval, a financial chart image including the second trading indicator of the trading day and the second trading indicator of the preceding trading day; and / or where the financial data includes financial report financial data, obtaining the financial report financial data in the following manner: for each financial reporting period within the first historical period of the company, obtaining, based on the company's financial report in the financial reporting period, the company's financial report indicators for the financial reporting period, thereby obtaining financial report financial data including the company's financial report indicators for each financial reporting period; generating a financial chart image of a target financial product based on the financial data, including: for each financial report indicator of the company within the first historical period, generating a financial chart image representing the financial report indicator within the first historical period, and / or for each financial report indicator in the latest financial reporting period of the company within the first historical period, generating a financial chart image representing the year-on-year comparison of the financial report indicator in the latest financial reporting period relative to the same financial reporting indicator in the previous year.

[0264] In this way, by pre-processing the obtained multi-dimensional financial data such as cleaning, sorting and analysis, the financial information can be further refined, and more intuitive and detailed financial information can be provided to users, further enabling users to efficiently obtain multi-dimensional financial information and improve the efficiency of users in obtaining financial information.

[0265] In one embodiment of the present invention, see Figure 7 A schematic diagram of the structure of another video push device is provided, which is applied to a client and includes:

[0266] The second target information obtaining module 701 is used to obtain the identifier of the target financial product and the target trading date input by the user;

[0267] The target information sending module 702 is used to send the identifier of the target financial product and the target trading date to the server, so that the server generates a target video according to any of the above-mentioned video push devices and feeds the target video back to the client;

[0268] The target video display module 703 is used to receive the target video and display the target video.

[0269] The embodiment of the present invention further provides an electronic device, such as Figure 8 As shown, it includes a processor 801 , a communication interface 802 , a memory 803 and a communication bus 804 , wherein the processor 801 , the communication interface 802 and the memory 803 communicate with each other via the communication bus 804 .

[0270] Memory 803, used for storing computer programs;

[0271] The processor 801 is configured to execute the program stored in the memory 803, and implement the following steps:

[0272] The communication bus mentioned in the electronic devices mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into address buses, data buses, control buses, etc. For ease of illustration, only a single thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0273] The communication interface is used for communication between the above electronic device and other devices.

[0274] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0275] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0276] In another embodiment of the present invention, a computer-readable storage medium is provided, in which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned video pushing methods are implemented.

[0277] In another embodiment of the present invention, a computer program product including instructions is provided. When the computer program product is run on a computer, the computer executes any one of the video pushing methods in the above embodiments.

[0278] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in accordance with the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disk, hard disk, tape), optical media (e.g., DVD), or semiconductor media (e.g., solid-state disk (SSD)).

[0279] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0280] Each embodiment in this specification is described in a related manner. Similar portions between the various embodiments can be referenced to each other. Each embodiment focuses on the differences between the other embodiments. In particular, the device, electronic device, storage medium, and computer program product embodiments are generally similar to the method embodiments, so their descriptions are relatively simple. For related portions, reference can be made to the descriptions of the method embodiments.

[0281] The above description is only a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention are included in the scope of protection of the present invention.

Claims

1. A video push method, characterized in that: Applied to the server, the method includes: Obtain the identifier of the target financial product and the target trading date entered by the user in the client; Based on the identifier and the target trading date, obtaining financial data of the target financial product; generating a financial chart image of the target financial product based on the financial data; Obtain news and public opinion data related to the target financial product; Converting the news and public opinion data into text data in a preset data format; generating an audio segment of speech that reads the text data; generating a video frame of a target video based on the duration of the audio segment and the financial chart image; Synchronizing the audio track of the audio segment with the video frame of the target video to obtain the target video, and feeding back the target video to the client; The step of generating a video frame of a target video based on the duration of the audio segment and the financial chart image includes: Obtaining an information content level of a material used to generate a video frame of a target video, wherein the material includes at least one of the following information: the financial chart image, the text data, and an image generated based on the text data, and the information content level represents the amount of financial information contained in the material; Determine the layer of each material in the video frame and the layout information of each material in the layer based on the information content level of each material; Generate a video frame of the target video based on the layer where each material is located, layout information of the material in the layer, and the duration of the audio segment; The information level of the material used to generate the video frame of the target video is obtained as follows: Obtaining an information content representation value of the material based on the financial data or news and public opinion data used to generate the material; determining the information content level of the material based on the information content representation value of the material, wherein the information content representation value of the material is determined in the following manner: in the case where the data used to generate the material is news and public opinion data, determining the timeliness representation value of the news and public opinion data based on the release time of the news and public opinion data, a preset timeliness reinforcement coefficient, and a preset time decay coefficient; obtaining the information content representation value of the material based on the word frequency representation value of each financial keyword in the news and public opinion data, the amount of text data in the news and public opinion data, and the timeliness representation value; the financial keywords in the news and public opinion data are: financial keywords obtained by keyword matching the news and public opinion data based on the financial keywords in a preset financial entity vocabulary; the word frequency representation value of the financial keywords is: a numerical value representing the word frequency of the financial keywords in the news and public opinion data and the representativeness of the financial keywords to the news and public opinion data; When the data used to generate the material is the intraday financial data and / or interday financial data of the target financial product, the information content representation value of the material is determined according to the following expression: ； in, The number of time units included in the quantity and price financial data. The quantity and price financial data includes: intraday financial data or interday financial data. is the trading volume of the t-th time unit, is the rate of return of the t-th time unit, It is the maximum value of the transaction volume in each time unit in the quantity and price financial data. The number of times the quantity and price financial data meets the preset quantity and price change trend; Determine the layout of each material in the target video as follows: For each material, determine the layout information of the material in the video frame according to the information content score of the material, the preset resolution of the target video, the preset horizontal distribution factor and the vertical amplitude coefficient; Adjust the main color values of the material as follows: According to the color information of the material, the color of the background and the information quantity representation value of the material, the visual effect color compensation value of the material is determined, and the main color value of the material is adjusted based on the visual effect color compensation value of the material.

2. The method according to claim 1, characterized in that After obtaining news and public opinion data related to the target financial product, it also includes: Segmenting the news and public opinion data; The step of converting the news and public opinion data into text data in a preset data format includes: For each paragraph of news and public opinion data obtained by segmentation, convert the news and public opinion data of the paragraph into text data in a preset data format; The step of generating an audio segment for broadcasting the text data includes: For each paragraph of text data obtained, generate an audio segment of the voice that broadcasts the text data of the paragraph; The step of generating a video frame of a target video based on the duration of the audio segment and the financial chart image includes: For each paragraph of news and public opinion data, generate a background image for that paragraph based on the text data of that paragraph; Generate video frames of a background video layer of a target video based on the duration of the audio segment of each paragraph and the background image of each paragraph; Based on the total duration of each audio segment and the financial chart image, a video frame of the information video layer of the target video is generated to obtain the video frame of the target video, wherein the picture of the information video layer is displayed on the picture of the background video layer.

3. The method according to claim 2, characterized in that The step of generating a video frame of a background video layer of a target video based on the duration of the audio segment of each paragraph and the background image of each paragraph includes: For each paragraph, generating an entry special effect video frame and an exit special effect video frame for the background image of the paragraph, and generating a video frame showing the background image of the paragraph between the entry special effect video frame and the exit special effect video frame to obtain segmented video frames of the paragraph, wherein the video duration of the generated video frame showing the background image of the paragraph is the same as the duration of the audio segment of the paragraph; The segmented video frames of each generated paragraph are spliced together to obtain the video frames of the background video layer of the target video.

4. The method according to claim 3, characterized in that The step of synchronizing the audio segments with the video frames of the target video to obtain the target video includes: Adding interval audio of an audio interval time between the audio segments of two adjacent paragraphs, wherein the audio interval time is equal to the sum of the exit special effect duration of the first paragraph and the entrance special effect duration of the second paragraph, the first paragraph is: the paragraph to which the segmented video frame played first in the video playback order of the two adjacent paragraphs belongs, and the second paragraph is: the paragraph to which the segmented video frame played later in the video playback order of the two adjacent paragraphs belongs.

5. The method according to any one of claims 1 to 4, characterized in that The financial data includes at least one of the following data: intraday financial data of the target financial product on the target trading day, interday financial data of the target financial product during a trading day interval, and financial data from financial reports of companies related to the target financial product, wherein the trading day interval is a first preset number of consecutive trading days including the target trading day; and / or In the case where the financial data includes the intraday financial data, the intraday financial data is obtained in the following manner: for each first preset time period in the target trading day, based on the trading data of the target financial product in the first preset time period, obtaining a first trading indicator of the target financial product for the first preset time period, thereby obtaining intraday financial data including the first trading indicator for each first preset time period in the target trading day; generating a financial chart image of the target financial product based on the financial data includes: for each first preset time period in the target trading day, generating, in chronological order of each first preset time period in the target trading day, a financial chart image including the first trading indicator for the first preset time period and the first trading indicators for the preceding time period of the first preset time period; and / or In the case where the financial data includes the intraday financial data, the intraday financial data is obtained in the following manner: for each trading day in the trading day interval, based on the trading data of the target financial product on that trading day, a second trading indicator of the target financial product on that trading day is obtained, thereby obtaining intraday financial data including the second trading indicator of each trading day in the trading day interval; and generating a financial chart image of the target financial product based on the financial data includes: for each trading day in the trading day interval, generating, in chronological order of each trading day in the trading day interval, a financial chart image including the second trading indicator of that trading day and the second trading indicator of the preceding time of that trading day; and / or In the case where the financial data includes the financial report financial data, the financial report financial data is obtained in the following manner: for each financial reporting period within the first historical time of the company, based on the financial report of the company in the financial reporting period, the financial report indicators of the company in the financial reporting period are obtained, and financial report financial data including the financial report indicators of the company in each financial reporting period are obtained; the financial chart image of the target financial product is generated based on the financial data, including: for each financial report indicator of the company within the first historical time, a financial chart image representing the financial report indicator within the first historical time is generated, and / or for each financial report indicator of the latest financial reporting period of the company within the first historical time, a year-on-year financial chart image representing the financial report indicator of the latest financial reporting period relative to the financial report indicator of the same financial reporting period in the previous year is generated.

6. A video push method, characterized in that: Applied to a client, the method includes: Obtaining the identifier of the target financial product and the target trading date input by the user; Sending the identifier of the target financial product and the target trading date to the server, so that the server generates a target video according to the video push method according to any one of claims 1 to 5, and feeds back the target video to the client; The target video is received and displayed.

7. A video push device, characterized in that: Applied to the server, the device includes: A first target information obtaining module is used to obtain the identifier of the target financial product and the target trading date input by the user on the client; a financial data acquisition module, configured to acquire financial data of the target financial product based on the identifier and the target trading date; a financial chart image generation module, configured to generate a financial chart image of the target financial product based on the financial data; A news and public opinion data acquisition module, used to obtain news and public opinion data related to the target financial product; A text data conversion module, used to convert the news and public opinion data into text data in a preset data format; An audio segment generating module, configured to generate an audio segment for broadcasting the text data; a video frame generation module, configured to generate a video frame of a target video based on the duration of the audio segment and the financial chart image; a target video feedback module, configured to synchronize the audio segments with the video frames of the target video to obtain the target video, and feed back the target video to the client; The video frame generation module is specifically configured to obtain an information content level of a material used to generate a video frame of a target video, wherein the material includes at least one of the following information: a financial chart image, text data, and an image generated based on the text data, and the information content level represents the amount of financial information contained in the material; based on the information content level of each material, determine the layer of each material in the video frame and the layout information within the layer; and generate a video frame of the target video based on the layer of each material, the layout information of the material within the layer, and the duration of the audio segment; The information level of the material used to generate the video frame of the target video is obtained as follows: Obtaining an information content representation value of the material based on the financial data or news and public opinion data used to generate the material; determining the information content level of the material based on the information content representation value of the material, wherein the information content representation value of the material is determined in the following manner: in the case where the data used to generate the material is news and public opinion data, determining the timeliness representation value of the news and public opinion data based on the release time of the news and public opinion data, a preset timeliness reinforcement coefficient, and a preset time decay coefficient; obtaining the information content representation value of the material based on the word frequency representation value of each financial keyword in the news and public opinion data, the amount of text data in the news and public opinion data, and the timeliness representation value; the financial keywords in the news and public opinion data are: financial keywords obtained by keyword matching the news and public opinion data based on the financial keywords in a preset financial entity vocabulary; the word frequency representation value of the financial keywords is: a numerical value representing the word frequency of the financial keywords in the news and public opinion data and the representativeness of the financial keywords to the news and public opinion data; When the data used to generate the material is the intraday financial data and / or interday financial data of the target financial product, the information content representation value of the material is determined according to the following expression: ； in, The number of time units included in the quantity and price financial data. The quantity and price financial data includes: intraday financial data or interday financial data. is the trading volume of the t-th time unit, is the rate of return of the t-th time unit, It is the maximum value of the transaction volume in each time unit in the quantity and price financial data. The number of times the quantity and price financial data meets the preset quantity and price change trend; Determine the layout of each material in the target video as follows: For each material, determine the layout information of the material in the video frame according to the information content score of the material, the preset resolution of the target video, the preset horizontal distribution factor and the vertical amplitude coefficient; Adjust the main color values of the material as follows: According to the color information of the material, the color of the background and the information quantity representation value of the material, the visual effect color compensation value of the material is determined, and the main color value of the material is adjusted based on the visual effect color compensation value of the material.

8. A video push device, characterized in that: Applied to a client, the device includes: A second target information obtaining module is used to obtain the identifier of the target financial product and the target trading date input by the user; a target information sending module, configured to send the identifier of the target financial product and the target trading date to the server, so that the server generates a target video according to the video push device according to claim 7 and feeds the target video back to the client; The target video display module is used to receive the target video and display the target video.

9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory for storing computer programs; A processor, configured to implement the method described in any one of claims 1 to 5 or 6 when executing a program stored in a memory.

Images