Live broadcast copywriting matching method, system and device, and storage medium

Abstract

The invention relates to the technical field of network live broadcast, and discloses a live broadcast copywriting matching method, system and device, and a storage medium. The method comprises the following steps: acquiring a to-be-live-broadcast product image, extracting to-be-live-broadcast product features in the to-be-live-broadcast product image, and generating to-be-live-broadcast product copywriting corresponding to the to-be-live-broadcast product features; acquiring a live video, and judging whether a copywriting matching action exists in the live video; if yes, extracting current product features of a current live product from the live video; judging whether the current product feature exists in the to-be-live-broadcast product features or not; and if yes, searching a current product copywriting corresponding to the current product feature from the to-be-live-broadcast product copywriting and outputting the current product copywriting. According to the application, synchronous pushing of the product copywriting and the live broadcast content can be realized, and the live broadcast cost is saved.

Description

Technical Field

[0001] The present application relates to the field of network live broadcast technology, and in particular to a live broadcast copy matching method, system, device and storage medium. Background Art

[0002] With the continuous development of Internet technology, the live broadcast industry has shown a blowout trend. Live broadcast has gradually developed into a national activity. A mobile phone that can connect to the Internet can initiate a simple live broadcast. In the field of live broadcasting, the anchor needs to explain the products in detail. The use of a teleprompter to display product information and other introductory text to the anchor has greatly reduced the burden on the anchor. However, in the uninterrupted live broadcast, a live assistant is needed to update the text of the teleprompter in real time, which increases the cost of live broadcasting. Summary of the invention

[0003] In a first aspect, the present application provides a live broadcast copy matching method, comprising:

[0004] Acquire an image of a product to be broadcast live, extract features of the product to be broadcast live in the image of the product to be broadcast live, and generate a text of the product to be broadcast live corresponding to the features of the product to be broadcast live;

[0005] Obtain a live video, and determine whether there is a text matching action in the live video;

[0006] If so, extracting the current product features of the current live product from the live video;

[0007] Determine whether the current product feature exists in the product features to be broadcast live;

[0008] If it exists, the current product copy corresponding to the current product features is searched from the product copy to be broadcast live and outputted.

[0009] In one implementation, determining whether there is a text matching action in the live video includes:

[0010] Acquire the hand gestures of the live broadcaster from the live broadcast video;

[0011] Through the gesture recognition model, detect whether the gesture action is one of the text matching actions; the text matching actions include the live broadcast personnel holding up the current live broadcast product with both hands, holding up the current live broadcast product with one hand and pointing at the current live broadcast product with the other hand, or pointing at the current live broadcast product with both hands.

[0012] In one embodiment, extracting the current product features of the current live product from the live video includes:

[0013] If the text-matching action is detected, obtaining an action recognition frame of the text-matching action from the live video;

[0014] Expanding the action recognition frame to obtain a product feature recognition frame, and using the product feature recognition frame to capture a live image from the live video;

[0015] Determine whether the current live broadcast product exists in the live broadcast image;

[0016] If so, extract the current product features corresponding to the current live broadcast product.

[0017] In one embodiment, the method further comprises:

[0018] If the current product feature is not one of the product features to be broadcast live, the step of obtaining the gestures of the live broadcaster from the live broadcast video is re-executed.

[0019] In one embodiment, the method further comprises:

[0020] A feature-copyright mapping database is constructed based on the features of the product to be broadcast live and the corresponding copy of the product to be broadcast live.

[0021] In one embodiment, the method further comprises:

[0022] If a new feature of a product to be broadcast live is detected, the new feature of the product to be broadcast live is matched with the existing product features in the feature copy mapping database;

[0023] If the existing product features contain target features that are the same as the new product features to be broadcast live, the target features and the corresponding target text are deleted, and a new product text is constructed and stored based on the new product features to be broadcast live.

[0024] In one embodiment, the method further comprises:

[0025] If the existing product features do not contain the same target features as the new product features to be broadcast live, a new product copy is constructed and stored according to the new product features to be broadcast live.

[0026] In a second aspect, the present application provides a live broadcast copy matching system, including: a client, a server, and a display;

[0027] The client is used to obtain the product image to be broadcast live;

[0028] The server end is used to extract the features of the product to be broadcast live in the product image to be broadcast live, and generate a product copy to be broadcast live corresponding to the features of the product to be broadcast live; obtain a live video, and determine whether there is a copy matching action in the live video; if so, extract the current product features of the current live product from the live video; determine whether there is the current product feature in the features of the product to be broadcast live; if so, search for the current product copy corresponding to the current product feature from the product copy to be broadcast live and output it to the display end;

[0029] The display end is used to display the current product copy corresponding to the current product feature.

[0030] In a third aspect, the present application provides a computer device, comprising a processor and a memory, wherein the memory stores a computer program, and the processor is used to execute the computer program to implement the live broadcast copy matching method described in the first aspect.

[0031] In a fourth aspect, the present application provides a computer storage medium storing a computer program, which, when executed on a processor, implements the live broadcast copy matching method according to the first aspect.

[0032] The embodiments of the present application have the following beneficial effects:

[0033] A live broadcast copy matching method provided in the present application can determine in real time whether copy matching is needed based on the live broadcast video, and when copy matching is needed, find the corresponding current product copy from the existing product copy to be broadcast and output it, thereby realizing the synchronous push of product copy and live broadcast content and saving live broadcast costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solution of the present application, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present application and should not be regarded as limiting the scope of protection of the present application. For ordinary technicians in this field, other related drawings can also be obtained based on these drawings without creative work.

[0035] Figure 1 A schematic diagram of a flow chart of a live broadcast copy matching method provided in this embodiment is shown;

[0036] Figure 2 A schematic diagram of a process flow of a text matching action detection method provided by this embodiment is shown;

[0037] Figure 3 A schematic diagram of a process flow of a current product feature extraction method provided in this embodiment is shown;

[0038] Figure 4 A structural diagram of a live broadcast copy matching system provided in this embodiment is shown. DETAILED DESCRIPTION

[0039] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments.

[0040] The components of the embodiments of the present application generally described and shown in the drawings herein may be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0041] Hereinafter, the terms "including", "having" and their cognates, which may be used in various embodiments of the present application, are intended only to indicate specific features, numbers, steps, operations, elements, components, or combinations of the foregoing items, and should not be understood as first excluding the existence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing items or adding the possibility of one or more features, numbers, steps, operations, elements, components, or combinations of the foregoing items.

[0042] Furthermore, the terms “first”, “second”, “third”, etc. are merely used for distinguishing descriptions and are not to be understood as indicating or implying relative importance.

[0043] Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meanings as those generally understood by those skilled in the art to which the various embodiments of the present application belong. The terms (such as those defined in generally used dictionaries) will be interpreted as having the same meanings as the contextual meanings in the relevant technical field and will not be interpreted as having idealized meanings or overly formal meanings unless clearly defined in the various embodiments of the present application.

[0044] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0045] Reference Figure 1 , Figure 1 A flowchart of a live copy matching method provided in this embodiment, which can realize synchronous push of product copy and live content, includes:

[0046] S101, obtaining an image of a product to be broadcast live, extracting features of the product to be broadcast live in the image of the product to be broadcast live, and generating a text of the product to be broadcast live corresponding to the features of the product to be broadcast live.

[0047] Before live broadcast, you can first determine all the products that need to be displayed in the subsequent live broadcast, and then collect images of all the products to be broadcast, such as taking product images with a camera. In order to improve the accuracy of subsequent product feature matching, the external environment when taking product images, such as lighting and brightness, can be kept as similar as possible to the environment during live broadcast.

[0048] Then, the features of the products to be broadcasted corresponding to each image of the products to be broadcasted are extracted. Since information unrelated to the products may be captured when the product images are captured, the images can be cropped after shooting to retain only the images of the product parts, and then the features can be extracted. Then, corresponding texts for the products to be broadcasted are generated or configured for the extracted features of the products to be broadcasted.

[0049] For example, if the products to be broadcast live include: mobile phones, induction cookers, and humidifiers, you can first obtain the images of the mobile phone, induction cooker, and humidifier respectively, and then extract the image features to obtain the image features of the mobile phone, the image features of the induction cooker, and the image features of the humidifier, and then assign the corresponding product copy. The product copy can be the price, function, and advantages of the product, etc., which can be set according to the needs of the live broadcast. The product copy can be automatically generated according to the copy generation tool, for example, by entering certain keywords and key parameters of the product in the copy generation tool to automatically generate the product copy. Alternatively, the user sets the product copy for each product to be broadcast live by himself, and then assigns it to the corresponding product. Ultimately, each image feature of the product to be broadcast live has a corresponding product copy, that is, a mapping relationship is formed between the product and the copy, and then a copy mapping database can be constructed to save the image features and product copy of the product to be broadcast live.

[0050] S102: Acquire a live video and determine whether there is a text matching action in the live video.

[0051] When live broadcasters are doing live broadcasts with goods, they usually perform various actions, such as picking up products, putting down products, introducing products, etc. For the continuity of the live broadcast, when the live broadcasters change the products, they usually do not manually change the product introduction copy. The traditional practice is that the assistants change the product introduction copy in real time according to the progress of the live broadcast, which undoubtedly increases the cost of the live broadcast.

[0052] Determining whether the product copy needs to be changed based on the live broadcaster's actions can ensure the continuity of the live broadcast and reduce the original copy of the live broadcast. Therefore, when the live broadcaster is broadcasting, the live video frame is obtained in real time, and then it is determined whether there is a copy matching action in the live video frame to determine whether the product copy needs to be changed.

[0053] S103: If so, extract the current product features of the current live product from the live video.

[0054] If there is a copy matching action, it means that the corresponding product copy needs to be matched for the currently live product. At this time, the current product features of the currently live product can be extracted from the live video frame, and then the corresponding product copy can be matched according to the current product features.

[0055] S104: Determine whether the current product feature exists in the product features to be broadcast live.

[0056] Because the product features of the product to be broadcast live have been extracted and the corresponding product text to be broadcast live has been generated before the live broadcast, it is only necessary to determine whether the current product feature of the current live product is one of the product features to be broadcast live.

[0057] S105. If it exists, search for the current product copy corresponding to the current product features from the product copy to be broadcast live and output it.

[0058] If the current product feature of the current live product is one of the features of the product to be live, then the current product copy corresponding to the current product feature can be directly searched from the product copy to be live and output.

[0059] This embodiment can determine in real time whether text matching is needed based on the live video, and when text matching is needed, find the corresponding current product text from the existing product texts to be broadcast live and output it, thereby realizing the synchronous push of texts with the live content and saving the live broadcast cost.

[0060] Reference Figure 2 , step S102 includes: steps S1021-S1022.

[0061] S1021. Acquire gestures of the live broadcaster from the live broadcast video.

[0062] S1022: Detect, through a gesture recognition model, whether the gesture action is one of the text matching actions.

[0063] Since the live broadcast personnel make many body movements during the live broadcast, most of the movements cannot be used to determine whether the current live broadcast product needs to match the corresponding product copy. When the live broadcast personnel introduce the product, their movements are mostly similar or the same, for example, the live broadcast personnel hold the current live broadcast product with both hands, hold the current live broadcast product with one hand and point to the current live broadcast product with the other hand, or point to the current live broadcast product with both hands.

[0064] Therefore, the copy matching action can be one of the above actions, or, a corresponding copy matching action can be set according to the habits or needs of the live broadcast personnel. As long as the live broadcast personnel performs the copy matching action during the live broadcast, the copy matching of the current live broadcast product will be started.

[0065] When performing gesture recognition on live broadcast personnel, a corresponding gesture recognition model may be used.

[0066] For example, images containing the text-matching actions of the live broadcast personnel can be used as training sets and annotated in advance. In order to improve the accuracy of the model, other images that do not contain text-matching actions, or images containing actions that are similar to the text-matching actions, can also be added to the training set.

[0067] Then, an existing deep learning model, such as the YOLO9 model, is used for training. When the training accuracy of the model reaches the expected value, such as the recognition accuracy rate exceeds 95%, the training is stopped, and then the trained gesture recognition model is used for gesture recognition during live broadcast.

[0068] This embodiment uses a gesture recognition model to determine whether the live broadcast personnel's gesture action is one of the copy matching actions, thereby improving the accuracy of action recognition. Moreover, gesture recognition is used to determine whether product copy matching is required later, and other personnel are not required to manually adjust the product copy, thereby ensuring the continuity of the live broadcast and reducing the cost of the live broadcast.

[0069] Reference Figure 3 , step S103 includes: steps S1031-S1034.

[0070] S1031. If the text matching action is detected, an action recognition frame of the text matching action is obtained from the live video.

[0071] S1032: Expand the action recognition frame to obtain a product feature recognition frame, and use the product feature recognition frame to capture a live image from the live video.

[0072] S1033: Determine whether the current live broadcast product exists in the live broadcast image.

[0073] S1034. If so, extract the current product features corresponding to the current live broadcast product.

[0074] Since the copy matching action is to detect the action of the live broadcaster, it is necessary to exclude the image of the product part during detection to avoid the image of the product part from affecting the recognition result of the copy matching action. Therefore, the action recognition frame corresponding to the copy matching action is also centered on the hands or other parts of the person being detected.

[0075] After confirming that the copy matching action is detected, it is necessary to extract the image of the product. When the copy matching action is the live broadcast personnel holding the product with both hands, holding the product with one hand, or pointing at the product, the live broadcast personnel's hands are often close to the current live broadcast product. Therefore, the product feature recognition frame can be constructed based on the action recognition frame.

[0076] For example, the action recognition frame is expanded five centimeters in all directions as a product feature recognition frame, or multiple circular product feature recognition frames with a radius of five centimeters are generated with the action recognition frame as the center, and the distance between each product recognition frame and the action recognition frame is less than two centimeters, and then the live video frame is intercepted through the product recognition frame. If the current live product exists in the screenshot image, the current product features of the current live product are extracted. If not, the next live video frame is intercepted. If the current live product is still not intercepted when a new copy matching action is detected, it means that the live broadcast personnel may have performed the copy matching action incorrectly. However, since the live broadcast device will not replace or cancel the old product copy when the current live product is not detected, the fault tolerance of the live broadcast is improved.

[0077] This embodiment generates a product feature recognition frame based on the action recognition frame, avoiding product recognition for the entire live video frame, reducing the workload of product feature recognition, and saving live broadcast costs.

[0078] In one embodiment, the method further comprises:

[0079] If the current product feature is not one of the product features to be broadcast live, the step of obtaining the gestures of the live broadcaster from the live broadcast video is re-executed.

[0080] During the live broadcast, the live broadcast staff may mistakenly issue a text matching action, such as pointing both hands in a direction other than the live broadcast product, or pointing at the wrong object. Then, when the current product features are subsequently extracted, the features of the wrong object may be extracted. Since the wrong object is not one of the products to be broadcast, it cannot be matched to the features of the product to be broadcast. At this time, the live broadcast staff's gestures need to be re-recognized so that the corresponding product copy can be matched normally for the live broadcast product when the live broadcast staff issues a text matching action next time.

[0081] In this embodiment, when the current product feature is not one of the product features to be broadcast live, it is determined that the live broadcaster has erroneously issued a text matching action, and the steps of gesture recognition and action acquisition are re-executed, thereby improving the fault tolerance rate of the live broadcast and ensuring the normal progress of the live broadcast.

[0082] In one embodiment, the method further comprises:

[0083] If a new feature of a product to be broadcast live is detected, the new feature of the product to be broadcast live is matched with the existing product features in the feature copy mapping database;

[0084] If the existing product features contain target features that are the same as the new product features to be broadcast live, the target features and the corresponding target text are deleted, and a new product text is constructed and stored based on the new product features to be broadcast live.

[0085] During the live broadcast process, the live broadcast staff may face various situations, such as being temporarily assigned a new product to carry, or changing the corresponding product copy for the stored product to be broadcast. At this time, the product features corresponding to the new product or the product whose product copy needs to be changed can be obtained, and then the new product features to be broadcasted are matched with the stored existing product features.

[0086] If the existing product features do not contain the same target features as the new product features to be broadcast live, it means that the product is a temporarily added product. At this time, it is necessary to construct and store a new product copy based on the new product features to be broadcast live.

[0087] If the existing product features contain the same target features as the new product features to be broadcast live, it means that the product needs to modify the product copy. At this time, the target features and the corresponding target copy need to be deleted, and a new product copy should be constructed and stored based on the new product features to be broadcast live.

[0088] When detecting new features of a product to be broadcast live, this embodiment matches the new features of the product to be broadcast live with the features of existing products in the copy mapping database, and takes different measures according to the matching results to meet the live broadcast needs in different situations.

[0089] Reference Figure 4 This embodiment also provides a live broadcast copy matching system 400, including: a client 401, a server 402 and a display 403.

[0090] The client 401 is used to obtain the product image to be broadcast live.

[0091] The server end 402 is used to extract the features of the product to be broadcast live in the product image to be broadcast live, and generate a product copy to be broadcast live corresponding to the features of the product to be broadcast live; obtain a live video, and determine whether there is a copy matching action in the live video; if so, extract the current product features of the current live product from the live video; determine whether the current product features exist in the features of the product to be broadcast live; if so, search for the current product copy corresponding to the current product features from the product copy to be broadcast live and output it to the display end 403.

[0092] The display terminal 403 is used to display the current product text corresponding to the current product features.

[0093] Among them, the display end can be a teleprompter or a display screen, and the live broadcast staff can view the current product copy corresponding to the current live broadcast product through the display end.

[0094] It can be understood that the system of this embodiment corresponds to the live broadcast copy matching method of the above embodiment, and the optional items in the above embodiment are also applicable to this embodiment, so they will not be repeated here.

[0095] The present application also provides a computer device. Exemplarily, the computer device includes a processor and a memory, wherein the memory stores a computer program, and the processor runs the computer program to enable the computer device to execute the functions of each module in the above-mentioned live broadcast text matching method or the above-mentioned live broadcast text matching system.

[0096] Among them, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including a central processing unit (CPU), a graphics processing unit (GPU) and a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or at least one of other programmable logic devices, discrete gates or transistor logic devices, and discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., which can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application.

[0097] The memory may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. The memory is used to store a computer program, and the processor may execute the computer program accordingly after receiving an execution instruction.

[0098] The present application also provides a computer storage medium for storing the computer program used in the above-mentioned computer device. The computer storage medium may be a readable storage medium, or a non-volatile storage medium or a volatile storage medium. For example, the computer storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0099] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and structure diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in an alternative implementation, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the structure diagram and / or the flow diagram, and the combination of boxes in the structure diagram and / or the flow diagram, can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0100] In addition, the functional modules or units in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0101] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a smart phone, a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application.

[0102] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A live broadcast copy matching method, characterized in that: include: Acquire an image of a product to be broadcast live, extract features of the product to be broadcast live in the image of the product to be broadcast live, and generate a text of the product to be broadcast live corresponding to the features of the product to be broadcast live; Obtain a live video, and determine whether there is a text matching action in the live video; If so, extracting the current product features of the current live product from the live video; Determine whether the current product feature exists in the product features to be broadcast live; If it exists, the current product copy corresponding to the current product features is searched from the product copy to be broadcast live and outputted.

2. The live broadcast copy matching method according to claim 1, characterized in that: The determining whether there is a text matching action in the live video includes: Acquire the hand gestures of the live broadcaster from the live broadcast video; Through the gesture recognition model, detect whether the gesture action is one of the text matching actions; the text matching actions include the live broadcast personnel holding up the current live broadcast product with both hands, holding up the current live broadcast product with one hand and pointing at the current live broadcast product with the other hand, or pointing at the current live broadcast product with both hands.

3. The live broadcast copy matching method according to claim 2, characterized in that: The extracting the current product features of the current live product from the live video includes: If the text-matching action is detected, obtaining an action recognition frame of the text-matching action from the live video; Expanding the action recognition frame to obtain a product feature recognition frame, and using the product feature recognition frame to capture a live image from the live video; Determine whether the current live broadcast product exists in the live broadcast image; If so, extract the current product features corresponding to the current live broadcast product.

4. The live broadcast copy matching method according to claim 3, characterized in that: The method further comprises: If the current product feature is not one of the product features to be broadcast live, the step of obtaining the gestures of the live broadcaster from the live broadcast video is re-executed.

5. The live broadcast copy matching method according to claim 1, characterized in that: The method further comprises: A feature-copyright mapping database is constructed based on the features of the product to be broadcast live and the corresponding copy of the product to be broadcast live.

6. The live broadcast copy matching method according to claim 5, characterized in that: The method further comprises: If a new feature of a product to be broadcast live is detected, the new feature of the product to be broadcast live is matched with the existing product features in the feature copy mapping database; If the existing product features contain target features that are the same as the new product features to be broadcast live, the target features and the corresponding target text are deleted, and a new product text is constructed and stored based on the new product features to be broadcast live.

7. The live broadcast copy matching method according to claim 6, characterized in that: The method further comprises: If the existing product features do not contain the same target features as the new product features to be broadcast live, a new product copy is constructed and stored according to the new product features to be broadcast live.

8. A live broadcast copy matching system, characterized in that: include: Client, server and display; The client is used to obtain the product image to be broadcast live; The server is used to extract the features of the product to be broadcast live in the product image to be broadcast live, and generate a product copy to be broadcast live corresponding to the features of the product to be broadcast live; Acquire a live video, and determine whether there is a copy matching action in the live video; if so, extract the current product features of the current live product from the live video; Determine whether the current product feature exists in the product features to be broadcast live; if so, search for the current product copy corresponding to the current product feature from the product copy to be broadcast live and output it to the display end; The display end is used to display the current product copy corresponding to the current product feature.

9. A computer device, characterized in that: The computer device includes a processor and a memory, the memory stores a computer program, and the processor is used to execute the computer program to implement the live broadcast copy matching method described in any one of claims 1-7.

10. A computer storage medium, characterized in that: It stores a computer program, which, when executed on a processor, implements the live broadcast copy matching method according to any one of claims 1-7.

Images