HK40038257B - Multimedia data processing method and apparatus, electronic device, and storage medium - Google Patents

Description

Multimedia data processing methods, devices, electronic equipment, and storage media

Technical Field

This application relates to the field of artificial intelligence, and in particular to a method, apparatus, electronic device, and storage medium for processing multimedia data.

Background Technology

With the continuous development of science and technology, traditional text and multimedia content with text and pictures can no longer meet the needs of users. Multimedia content such as video and audio (such as short videos) has gradually become one of the important ways for the public to obtain information and entertainment.

In daily life, users often prefer to directly browse key information (such as highlights in short videos) to quickly obtain relevant information. To meet this need, existing technologies in artificial intelligence and big data often employ image-text matching between multimedia content and title information to suggest key media content related to the title. However, since images and text belong to different fields, the matching process often results in poor performance. Furthermore, existing image-text matching technologies cannot identify the key audio content related to the title information in audio data that does not contain any images, leading to poor applicability and a reduced user experience.

Therefore, accurately identifying the main content in multimedia data has become an urgent problem to be solved.

Summary of the Invention

This application provides a multimedia data processing method, apparatus, electronic device, and storage medium, which can determine the playback time range of the main content associated with the title information in the multimedia data, improve the user experience, and has high applicability.

In a first aspect, embodiments of this application provide a method for processing multimedia data, the method comprising:

Obtain at least one text information contained in the multimedia data, as well as the title information of the multimedia data;

Determine the degree of matching between the above title information and each of the above text information;

Based on the matching degree of each of the above text information, the target playback time area in the above multimedia data is determined;

The multimedia data is processed according to the target playback time range.

Secondly, embodiments of this application provide a multimedia data processing apparatus, the apparatus comprising:

The acquisition unit is used to acquire at least one text information contained in the multimedia data, as well as the title information of the multimedia data.

The determining unit is used to determine the degree of matching between the above-mentioned title information and each of the above-mentioned text information;

The aforementioned determining unit is used to determine the target playback time region in the aforementioned multimedia data based on the matching degree corresponding to each of the aforementioned text information.

The playback unit is used to process the multimedia data according to the target playback time range.

Thirdly, embodiments of this application provide an electronic device, including a processor and a memory, which are interconnected;

The aforementioned memory is used to store computer programs;

The processor is configured to execute the method provided in the first aspect when the computer program is invoked.

Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that is executed by a processor to implement the method provided in the first aspect above.

Fifthly, embodiments of this application provide a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method provided in the first aspect.

In this embodiment, by representing multimedia data with at least one text message, the matching degree between the title information and each text message can be accurately determined at the text level. Furthermore, the correlation between each text message and the title information can be accurately measured based on the matching degree to determine the target playback time zone in the multimedia data. Further, processing the multimedia data using the target playback time zone allows users to quickly determine the playback time zone of the multimedia content related to the title information, enhancing user appeal and improving applicability.

Attached Figure Description

To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

Figure 1 is a flowchart of a multimedia data processing method provided in an embodiment of this application;

Figure 2a is a schematic diagram of a scenario for obtaining text information contained in video data according to an embodiment of this application;

Figure 2b is a schematic diagram of another scenario for obtaining text information contained in video data according to an embodiment of this application;

Figure 2c is a schematic diagram of another scenario for obtaining text information contained in video data according to an embodiment of this application;

Figure 3 is a schematic diagram of a scenario for obtaining text information contained in audio data according to an embodiment of this application;

Figure 4 is a schematic diagram of determining the target playback time region based on similarity according to an embodiment of this application;

Figure 5 is a schematic diagram of determining the target playback time region based on keywords according to an embodiment of this application;

Figure 6 is a schematic diagram of determining the target playback time region based on specified information according to an embodiment of this application;

Figure 7 is a schematic diagram of determining the target playback time region provided in an embodiment of this application;

Figure 8 is a schematic diagram of a scenario for processing multimedia data provided in an embodiment of this application;

Figure 9 is a schematic diagram of the structure of the multimedia data processing device provided in an embodiment of this application;

Figure 10 is a schematic diagram of the structure of the electronic device provided in an embodiment of this application.

Detailed Implementation

The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

The multimedia data processing method provided in this application embodiment is applicable to various fields such as artificial intelligence and big data, including human-computer interaction based on natural language processing (NLP), cloud computing in cloud technology, artificial intelligence cloud services, and related data computing and processing fields in the big data field. It aims to convert multimedia data into text information and then determine the target playback time area of the main media content in the multimedia data based on the text information.

Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

Natural Language Processing (NLP) is an important field within computer science and artificial intelligence. It studies the theories and methods for enabling effective communication between humans and computers using natural language. NLP is a science that integrates linguistics, computer science, and mathematics. Therefore, research in this field involves natural language—the language people use in daily life—and thus it has a close relationship with linguistic research.

Cloud technology refers to a hosting technology that unifies hardware, software, network, and other resources within a wide area network (WAN) or local area network (LAN) to achieve data computation, storage, processing, and sharing. The multimedia data processing method provided in this application embodiment can be implemented based on cloud computing.

Cloud computing refers to obtaining the required resources on demand and in an easily scalable manner through the network. It is the product of the development and integration of traditional computer and network technologies such as grid computing, distributed computing, parallel computing, utility computing, network storage technologies, virtualization, and load balancing.

Artificial intelligence cloud services are generally also known as AIaaS (AI as a Service). This is currently a mainstream service model for artificial intelligence platforms. Specifically, AIaaS platforms break down several common artificial intelligence services and provide them as independent or packaged services in the cloud, such as speech recognition processing and text information extraction.

Big data refers to data sets that cannot be captured, managed, and processed within a certain timeframe using conventional software tools. It represents massive, rapidly growing, and diverse information assets that require new processing models to achieve stronger decision-making, insightful discovery, and process optimization capabilities. With the advent of the cloud era, big data has attracted increasing attention. This embodiment effectively implements the multimedia data processing method provided, based on technologies such as massively parallel processing databases, data mining, distributed file systems, distributed databases, and cloud computing.

Referring to Figure 1, Figure 1 is a flowchart of a multimedia data processing method provided in an embodiment of this application. This method can be executed by any electronic device, such as a server or a user terminal, or it can be completed through interaction between the user terminal and the server. When executed by a user terminal, after acquiring multimedia data, the user terminal can determine the target playback time region in the multimedia data, and then process the multimedia data based on the target playback time region. When completed through interaction between the server and the user terminal, the server can determine the target playback time region in the multimedia data, and then indicate the target playback time region to the user terminal, which then processes the multimedia data according to the target playback time region. The multimedia data received by the server can be sent by the user terminal, or it can be obtained by the server through other means, such as databases or web pages; no limitations are imposed here. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server or server cluster providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. User terminals can be smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, etc. User terminals and servers can be connected directly or indirectly through wired or wireless communication, but are not limited to these.

As shown in Figure 1, the multimedia data processing method provided in this application embodiment may include the following steps:

Step S101: Obtain at least one text information contained in the multimedia data, as well as the title information of the multimedia data.

In some feasible implementations, the multimedia data in the embodiments of this application includes, but is not limited to, video data, audio data, and data combining video and audio. The video data is multimedia data containing both image and audio data, and the audio data can be audio data from the video data. The specific details can be determined based on the actual application scenario and are not limited here. Furthermore, the embodiments of this application can determine the target playback time region in the multimedia data based on at least one text information contained within the multimedia data, so as to process the multimedia data based on the target playback time region.

In some feasible implementations, for video data, caption information from at least one frame of the video data can be obtained, and this caption information from at least one frame can be used as at least one piece of text information contained in the video data. That is, caption information from any frame of the video data can be used as a piece of text information contained in the video data. When obtaining caption information from the frame images of the video data, OCR (Optical Character Recognition) technology, or other text recognition methods, text extraction tools, etc., can be used, and there are no limitations on this method.

Referring to Figure 2a, which is a schematic diagram of a scenario for obtaining text information contained in video data according to an embodiment of this application. For ease of description, assume that a video data set has only 9 frames, and each frame has different subtitle information, such as the subtitle information in the first frame being "Typhoon Mangkhut is about to make landfall," and the subtitle information in the seventh frame being "Let's see what happened." For this video data, the subtitle information of each frame can be used as multiple pieces of text information contained in the video data. For example, the subtitle information "Typhoon Mangkhut is about to make landfall" in the first frame can be used as one piece of text information contained in the video data, and the subtitle information "Let's see what happened" in the seventh frame can be used as another piece of text information contained in the video data. Optionally, either the subtitle information in the first frame or the seventh frame can be used as one piece of text information according to actual needs. The specific use can be determined according to the actual application scenario requirements and is not limited here.

Specifically, due to the principle of visual persistence in the human eye—meaning the eye cannot distinguish single static images—to achieve a smooth and continuous visual playback effect, video data is played as a series of continuously changing images (e.g., 24 consecutive frames per second). In this case, it is common for multiple consecutive frames in the video data to contain identical subtitle information. Therefore, after identifying each frame in the video data, for multiple consecutive frames with identical subtitle information, the subtitle information of any one or more frames can be used as one or more pieces of text information contained in the video data. For example, the subtitle information of the first frame played in a series of consecutive frames with identical subtitle information can be used as one piece of text information contained in the video data.

Referring to Figure 2b, which is a schematic diagram of another scenario for obtaining text information contained in video data according to an embodiment of this application. For ease of description, assume there is video data containing 60 frames, where frames 1 to 35 all correspond to the same image, and the caption information is "Typhoon 'Mangkhut' is about to make landfall," and frames 36 to 60 all correspond to the image in Figure 1, and the caption information is "Let's see what happened." In this case, the caption information of any frame from frames 1 to 35 (such as frame 1) can be considered as one piece of text information contained in the video data, and the caption information of any frame from frames 36 to 60 (such as frame 36) can be considered as one piece of text information contained in the video data. That is, two pieces of text information contained in the video data can be determined through the above method.

Optionally, for video data where each character in the subtitle information appears sequentially as the video plays, using the text information corresponding to each frame as the text information included in the video data would result in a large amount of text information lacking semantic meaning or with incomplete semantic representation. Therefore, for this type of video data, the video data can be divided into multiple video data segments, such that each video data segment consists of all frame images corresponding to a complete subtitle line. That is, each video data segment is all frame images corresponding to a complete subtitle line from its first character to the complete subtitle line. Then, the frame image containing the complete subtitle can be determined from all frame images of each video data segment, and the subtitle information of that frame image can be used as at least one piece of text information included in the video data. Based on the above implementation, each complete subtitle line in the video data can be used as one piece of text information included in the video data, improving the processing efficiency of text information.

Referring to Figure 2c, which is another scenario illustration of obtaining text information contained in video data according to an embodiment of this application, Figure 2c shows a video data segment containing 9 frames. The subtitle information in each frame is a part of a complete subtitle. For example, the subtitle information in the first frame is "台" (Taiwan), and the subtitle information in the second frame is "台风" (Typhoon). As the video data segment is played continuously, the subtitle information changes with the playback screen until the complete subtitle information "Typhoon 'Mangkhut' is about to make landfall" is displayed. As shown in Figure 2c, since the subtitle information in frames 1 to 8 is not complete, only the subtitle information in frame 9 is complete. Therefore, for a video data segment like the one in Figure 2c, the subtitle information contained in frame 9 can be regarded as a piece of text information contained in the video data.

Optionally, for video data that does not contain subtitle information, if the video data contains audio data, such as video narration or news broadcast audio, the audio data can be subjected to speech recognition to obtain the speech recognition result. Then, the text content corresponding to each sentence in the speech recognition result can be used as at least one piece of text information contained in the video data.

In some feasible implementations, since audio data does not contain frame images, it is impossible to directly extract the text information contained in the audio data from the frame images. In this case, natural language processing techniques can be used to convert the audio data into text content, and then the at least one piece of text information contained in the audio data can be determined based on the corresponding text content. Specifically, speech recognition can be performed on the audio data to obtain the speech recognition result. The text content corresponding to each sentence in the speech recognition result can be used as at least one piece of text information contained in the audio data. That is, each sentence in the speech recognition result of the audio data can be used as a piece of text information contained in the audio data.

Referring to Figure 3, which is a schematic diagram of a scenario for obtaining text information contained in audio data according to an embodiment of this application. As shown in Figure 3, after performing speech recognition on the audio data, the obtained speech recognition result includes two statements: "Typhoon Mangkhut is about to make landfall" and "Let's see what happened." For the above two statements, the text content corresponding to each statement can be used as a piece of text information contained in the speech data.

Optionally, for each statement in the speech recognition result, when the text length of any statement is short, the semantics expressed by its corresponding text content are limited. Therefore, after obtaining the speech recognition result, the statements with short text lengths in the speech recognition result are removed, and the text content corresponding to each statement with a text length greater than a preset text length threshold can be used as a text information contained in the audio data.

Optionally, when there are sentences with the same text content in the speech recognition results of the audio data, the text content corresponding to the sentence with the earliest playback time can be used as a text information contained in the speech data, or the text content corresponding to each sentence can be used as a text information contained in the audio data. The specific method can be determined based on the actual application scenario requirements, and there are no restrictions here.

Optionally, since speech recognition results often contain some interjections and other meaningless words, after obtaining the speech recognition results of the audio data, the speech recognition results can be filtered to remove interjections and other meaningless words. Then, based on the filtered speech recognition results, at least one text information contained in the audio data can be determined according to any of the above feasible implementation methods.

Optionally, when the audio data is speech data contained in the video data, at least one text information contained in the video data can be determined based on the text content corresponding to at least one sentence in the speech recognition result of the audio data.

In some feasible implementations, the title information of multimedia data can be the file name of the multimedia data, the topic information associated with the multimedia data, and a brief description, such as the video title on a short video platform, or content tags about video and audio in blog content. The specific details can be determined based on the actual application scenario and are not limited here.

Step S102: Determine the matching degree between the title information and each of the text information.

In some feasible implementations, after obtaining the title information of the multimedia data, the matching degree between the title information and each text information in at least one text information can be determined, and then the target playback time area in the multimedia data can be determined based on the matching degree corresponding to each text information.

The matching degree between the title information and each piece of text information represents the degree of association between the title information and each piece of text information, thus characterizing the degree of association between the multimedia content corresponding to each piece of text information and the title information. Furthermore, the higher the matching degree between any piece of text information and the title information, the higher the degree of association between the multimedia content corresponding to that text information and the title information, and the closer the multimedia content corresponding to that text information is to the title information; that is, the multimedia content corresponding to that text information is the main content of the multimedia data.

In some feasible implementations, the matching degree between the title information and the text information contained in the multimedia data can be considered as the text similarity between the title information and the text information contained in the multimedia data. That is, the higher the text similarity between the title information and any text information, the higher the correlation between the multimedia content corresponding to that text information and the title information.

The text similarity between the title information and each text information can be determined by calculating the cosine similarity, Euclidean distance, Hamming distance, and Jaccard similarity between the title information and each text information. The specific method can be determined based on the actual application scenario and is not limited here.

In some feasible implementations, the matching degree between the title information and the various textual information contained in the multimedia data can also be determined based on the keywords in the title information. The keywords in the title information are words that represent the main information in the title information. For example, if the title information is "Opening dates for universities in multiple locations confirmed," the keywords in this title information could be "universities," "opening dates," and "confirmed." It should be noted that the specific method for determining the keywords in the title information can be determined according to the specific title information and the actual application scenario requirements, and is not limited here.

Specifically, when determining the matching degree of each text message based on the keywords in the title information, the keywords in the title information can be identified first. Then, each text message is further segmented to obtain all the words in it. For each text message, all its words can be matched with the keywords in the title information to determine the frequency of each keyword. For example, if "university" appears twice and "opening time" appears once, the higher the total frequency of all keywords in the title information within any text message, the stronger the correlation between the two. For instance, if one text message contains all keywords twice and another contains them eight times, the latter is clearly more correlated with the title information. Therefore, for each text message, the total frequency of all keywords is determined as the matching degree between the title information and the text message.

Optionally, while the keywords included in the title information can represent the main information, each keyword has a different level of importance within the title. For example, in the title information "Opening dates for universities in multiple regions confirmed," the keywords "universities," "opening dates," and "confirmed" have significantly higher importance than "confirmed." Therefore, using the frequency of each keyword in the text information as the matching degree will ignore the importance of each keyword to the title information, resulting in a situation where text information with high relevance to the title information but fewer occurrences of certain keywords corresponds to a lower matching degree. Therefore, in the above case, for each piece of text information, the frequency of each keyword can be determined. Then, based on the weight of each keyword, the sum of the weights of all keywords corresponding to the text information is obtained, and this sum of weights is determined as the matching degree between the title information and the text information.

In some feasible implementations, before determining the matching degree between the title information and each piece of text information, it can be first determined whether any text information contains specified information. If no text information containing the specified information exists, then the matching degree between the title information and each piece of text information is determined. The specified information refers to common phrases used to indicate the main content of the multimedia data, such as "Let's see what happens next" or "This episode mainly includes the following content," which are common in short videos. The specific text content of the specified information can be determined based on the actual application scenario requirements and is not limited here. In other words, when there is text information containing the specified information among the text information included in the multimedia data, the playback content corresponding to the text information containing the specified information is the main multimedia content of the multimedia data, thus determining that the text information has a strong correlation with the title information. Therefore, for any piece of text information, when it contains the specified information, it can be determined that the text information is related to the title information.

Furthermore, when no text information containing the specified information exists among the various text information, the text similarity between the title information and each text information can be determined first. If a text similarity that meets preset conditions exists, the text similarity corresponding to each text information can be determined as the matching degree corresponding to each text information. The aforementioned text similarity meeting preset conditions can include text similarity exceeding a text similarity threshold, or text similarity exceeding a certain number, etc., and can be determined based on the actual application scenario; no restrictions are imposed here.

When no text similarity matches the preset conditions in the text similarity scores of each text piece, the matching degree between the title information and each text piece can be determined based on the frequency of each keyword in the title information appearing in each text piece. Alternatively, the weighted sum of each text piece can be determined based on the frequency of each keyword in the title information appearing in each text piece, as well as the weight of each keyword, and then the weighted sum of each text piece can be used to determine the matching degree between the title information and each text piece.

Step S103: Determine the target playback time range in the multimedia data based on the matching degree of each text information.

In some feasible implementations, the target playback time region is the playback time region of the multimedia content associated with the title information in the multimedia data, or the playback time region of the main content in the multimedia data. For example, if the title information of a short video is "Car accident occurred on Xinhua Street," the target playback time region in the short video can be the playback time region of the video content corresponding to the scene of the car accident.

Specifically, when the multimedia data is video data, the playback time region corresponding to each text message is the playback time region of the frame image corresponding to that text message; when the multimedia data is audio data, the playback time region of the text message is the playback time region of the first character in the text message in the audio data.

Optionally, the start playback time of the frame image or audio data corresponding to each text information can be regarded as the playback time region corresponding to each text information. That is, the playback time region corresponding to each text information can represent a time region with a very small time span or represent the start playback time corresponding to the text information. The specific time region can be determined based on the actual application scenario requirements and is not limited here.

In some feasible implementations, since the matching degree of each text message can represent the degree of association with the title information, after determining the matching degree of each text message, the playback time area corresponding to the text message that meets the matching conditions can be determined as the target playback time area in the multimedia data. That is to say, when the degree of association between any text message and the title information reaches a certain level, it can be determined that the multimedia content corresponding to that text message is related to the title information.

The text information that meets the matching criteria can be either text information with a matching degree higher than the matching degree threshold, or text information with the highest matching degree among all text information. The specific criteria can be determined based on the actual application scenario and are not limited here. The matching degree threshold can also be determined based on the application scenario and is not limited here.

Optionally, when the matching degree corresponding to each text information is the text similarity between the title information and each text information, the playback time region corresponding to the text information whose text similarity satisfies the above matching conditions can be determined as the target playback time region in the multimedia data. Wherein, when the matching degree is text similarity, the matching degree threshold is the corresponding similarity threshold. See Figure 4, which is a schematic diagram of determining the target playback time region based on text similarity provided in this application embodiment. Assuming the text similarity between the title information and each text information is cosine similarity, when determining the text similarity between the title information and a certain text information, the title information and the text information can be vectorized to obtain the text information vector corresponding to the text information and the title information vector corresponding to the title information. Based on the text information vector corresponding to the text information and the title information vector corresponding to the title information, the cosine similarity between the title information and the text information is determined, and this is taken as the matching degree between the title information and the text information. Therefore, the playback time region corresponding to the text information that satisfies the matching conditions is determined as the target playback time region in the multimedia data.

Optionally, when the matching degree of each text information in the multimedia data is determined based on the frequency of each keyword in the title information appearing in each text information, the playback time area corresponding to the text information whose matching degree meets the matching condition can be determined as the target playback time area in the multimedia data. Refer to Figure 5, which is a schematic diagram of determining the target playback time area based on keywords according to an embodiment of this application. In Figure 5, for each text information, word segmentation can be performed to extract each word. Further, each word in the text information is matched with the keywords in the title information to determine the frequency of each keyword appearing in each text information. Then, based on the frequency of each keyword appearing in each text information, the matching degree corresponding to each text information is determined, so that the playback time area corresponding to the text information that meets the matching degree condition is determined as the target playback time area in the multimedia data.

Optionally, when the matching degree corresponding to each text message is the sum of weights determined based on the keywords appearing in the title information in each text message, the playback time area corresponding to the text message whose matching degree meets the matching condition can be determined as the target playback time area in the multimedia data. In this case, the matching condition is that the sum of weights is higher than the weight sum threshold, or the sum of weights is the highest weight sum.

In some feasible implementations, to avoid having too many target playback time regions determined based on the above implementation method, if the text information that meets the matching conditions includes at least two text information, then, among the text information that meets the matching conditions, one or more text information corresponding to the playback time region can be selected as the target playback time region in the multimedia data based on a preset selection method.

Optionally, the playback time corresponding to each text message that meets the matching conditions in the multimedia data can be determined, and the playback time area corresponding to the first or the first preset number of text messages that meet the matching conditions can be determined as the target playback time area in the multimedia data.

Optionally, from the text information that meets the matching conditions, a preset number of text information corresponding to the playback time area are randomly selected and determined as the target playback time area in the multimedia data.

Optionally, based on the playback time of each text message that meets the matching conditions in the multimedia data, the playback time corresponding to the first text message is determined as a target playback time region in the multimedia data. For any text message after the first text message, if the time distance between the playback time region corresponding to the text message and its adjacent previous playback time region is not less than the time distance threshold, then the playback time region corresponding to the text message can be determined as the target playback time region in the multimedia data.

It should be noted that the above implementation method for determining the target playback time region in multimedia data from text information that meets the matching conditions based on the preset selection method is only an example. The specific implementation method can be determined based on the actual application scenario, and there are no restrictions here.

In some feasible implementations, when the matching condition is that the matching degree is higher than the matching degree threshold, the matching degree of each text information in the multimedia data can be determined sequentially according to the playback time of each text information in the multimedia, following the playback order. After determining the matching degree of each text information, it is compared with the matching degree threshold. Thus, the playback time region corresponding to the first or a predetermined number of text information with a matching degree higher than the matching degree threshold is determined as the target playback time region in the multimedia data. Alternatively, after determining the playback time region corresponding to the first text information with a matching degree higher than the matching degree threshold as a target playback time region in the multimedia data, the next text information with a matching degree higher than the matching degree threshold can be determined. If the time distance between the playback time region corresponding to this text information and the playback time region of the first text information with a matching degree higher than the matching degree threshold is not less than the time distance threshold, then the playback time region corresponding to this text information can be determined as the target playback time region in the multimedia data, and so on, until all target playback time regions in the multimedia data are determined.

In some feasible implementations, when the matching degree between the title information and each text information is determined by the number of times each keyword in the title information appears in each text information, the total number of times all keywords appear in any text information can be determined as the matching degree corresponding to that text information. If the above matching condition is that the total number of occurrences is higher than the threshold, then when there are multiple text information that meet the matching condition (multiple text information with a total number of occurrences higher than the threshold), the weight sum of each text information is determined based on the weight of each keyword, and the playback time region corresponding to the text information with the highest weight sum is determined as the target playback time region. If there are multiple text information with the highest weight sum among the multiple text information that meet the matching condition, the playback time region corresponding to the first text information with the highest weight sum can be determined as the target playback time region in the multimedia data based on the playback time of each text information in the multimedia data. Alternatively, the target playback time region in the multimedia data can be determined from the multiple text information with the highest weight sum based on the above preset selection method, without any limitation.

Optionally, if the matching condition is the highest total number of occurrences, then when multiple text messages satisfy the matching condition (multiple text messages with the highest total number of occurrences) exist among the text messages satisfying the matching condition, the text message with the highest weight among the multiple text messages satisfying the matching condition is determined based on the number of times each keyword appears in each text message and the weight of each keyword, and the playback time area corresponding to the text message with the highest weight is determined as the target playback time area. If multiple text messages satisfying the matching condition have multiple text messages with the highest weight, the playback time area corresponding to the first text message with the highest weight can be determined as the target playback time area based on the playback time of each text message in the multimedia data. Alternatively, the target playback time area in the multimedia data can be determined from the multiple text messages with the highest weight based on the above preset selection method, without any restrictions.

In some feasible implementations, when the matching degree between the title information and each text information is the sum of the weights corresponding to each text information, if the matching condition is that the sum of weights is higher than the weight sum threshold, or the sum of weights is the highest weight sum, then when there are multiple text information pieces that satisfy the matching condition (multiple text information pieces with a sum of weights higher than the weight sum threshold, or multiple text information pieces with a sum of weights higher than the weight sum threshold), the playback time region corresponding to the text information piece with the highest total frequency of all keywords is determined as the target playback time region. If there are multiple text information pieces with the highest total frequency among the multiple text information pieces that satisfy the matching condition, the playback time region corresponding to the first text information piece with the highest total frequency can be determined as the target playback time region in the multimedia data based on the playback time of each text information piece in the multimedia data. Alternatively, the target playback time region in the multimedia data can be determined from the multiple text information pieces with the highest total frequency based on the above-mentioned preset selection method, without any restrictions.

In some feasible implementations, since the specified information in step S102 can indicate the main playback content of the multimedia data, the target playback time area in the multimedia data can be determined based on the specified information before determining the matching degree between the title information and each text information. Refer to Figure 6, which is a schematic diagram of determining the target playback time area based on specified information according to an embodiment of this application. After obtaining at least one text information contained in the multimedia data, each text information can be compared with the specified information, and the playback time area corresponding to the text information containing the specified information can be further determined as the target playback time area in the multimedia data. When determining the target playback time area in the multimedia data based on the specified information, one or more specified information can be used, depending on the actual application scenario, and no limitation is made here. For example, the playback time area corresponding to the text information containing any specified information can be determined as the target playback time area in the multimedia data.

Specifically, each text message can be matched with the specified information in sequence according to the playback time. During the matching process, the playback time area corresponding to the first or the first preset number of text messages containing any specified information is determined as the target playback time area in the multimedia data.

Optionally, all text information containing any specified information can be identified first, and then text information containing different specified information can be identified from among them. If multiple text information containing the same specified information exist, the text information with the earliest playback time is selected. In other words, based on the above implementation method, text information containing different specified information (the specified information contained in any text information is different from the specified information contained in other text information) is identified from all text information contained in the multimedia data, and then the playback time area corresponding to each text information containing different specified information is determined as the target playback time area in the multimedia data.

Further, referring to Figure 7, which is a schematic diagram of determining the target playback time region according to an embodiment of this application. In Figure 7, if each text information in the multimedia data contains text information containing specified information, the target playback time region in the multimedia data can be successfully determined. If no text information containing specified information exists in the text information, i.e., determining the target playback time region in the multimedia data based on the specified information fails, the text similarity between the title information and each text information can be determined. The text similarity is used as the matching degree corresponding to each text information, and the target playback time region in the multimedia data is further determined based on the matching conditions. The specific determination method is as described above and will not be repeated here. If the target playback time region in the multimedia data is not determined when using text similarity as the matching degree, i.e., determining the target playback time region based on text similarity fails, the number of times each keyword in the title information appears in each text information can be determined based on keyword matching, thereby determining the matching degree between the title information and each text information. The target playback time region in the multimedia data is then determined based on the matching conditions. The specific determination method is as described above and will not be repeated here.

It should be noted that the above-described specific implementation method for determining the target playback time region in multimedia data based on specified information is only an example. The specific implementation method can be determined based on the actual application scenario, and no restrictions are imposed here.

In some feasible implementations, the combination order of the three methods shown in Figure 7—namely, the method of determining the target playback time area by specifying information, the method of determining the target playback time area by using text similarity as the matching degree, and the method of determining the target playback time area based on the occurrence frequency of each keyword in the text information—is only an example. The specific combination order may be limited based on the actual application scenario and is not restricted here.

For example, the text similarity between the title information and each piece of text information can be determined first. This text similarity can then be used as a matching score to determine the target playback time region in the multimedia data based on matching criteria. If determining the target playback time region based on text similarity fails, the matching score of each piece of text information can be determined based on the frequency of each keyword appearing in the title information. Then, the target playback time region can be determined based on the matching criteria. Furthermore, if determining the target playback time region based on the frequency of each keyword appearing in each piece of text information fails, the target playback time region in the multimedia data can be determined based on the aforementioned specified information.

Optionally, any two of the above three methods for determining the target playback time area can be combined to obtain a new method for determining the target playback time area. The specific selection method and combination order can also be determined based on the actual application scenario, and there are no restrictions here.

For example, the text similarity between the title information and each piece of text information can be determined first. This text similarity can then be used as a matching score to determine the target playback time region in the multimedia data based on the matching criteria. If determining the target playback time region based on text similarity fails, the target playback time region in the multimedia data can be determined based on the aforementioned specified information.

Step S104: Process the multimedia data according to the target playback time range.

In some feasible implementations, the multimedia content corresponding to the target playback time area determined in step S102 is the main content associated with the title information of the multimedia data. Therefore, after determining the target playback time area in the multimedia data, playback prompt information can be generated based on the target playback time area to indicate the target playback time area of the main content of the multimedia data.

For short video applications, different users may have varying tolerance levels for the length of prelude (video content unrelated or only marginally relevant to the title) due to factors such as personality and environment. Therefore, playback prompts generated based on the target playback time range can indicate the target playback time range of the video's highlights (the main video content relevant to the title), satisfying users' need for quick viewing and improving user experience.

For music applications, the text information corresponding to the target playback time area is related to the title information. That is, the audio content corresponding to the target playback time area may be the chorus (climax) of the music. Therefore, music applications can prompt users with playback prompts to indicate the target playback area corresponding to the chorus of the music, so that users can directly enjoy the chorus of the music, or extract the chorus of the music based on the playback prompts (such as extracting the chorus as a ringtone), etc.

Specifically, when playback prompts are used to indicate the target playback time range of the main content of multimedia data, these prompts can be text, voice, symbols, graphics, such as symbols in a video playback progress bar, or frame images and text information corresponding to the target playback time range. The specific method can be determined based on the actual application scenario requirements and is not limited here. Furthermore, during the playback of multimedia data, such as when playing video or audio content, playback prompts can be displayed to the user to indicate the target playback time range of the main content related to the title information of the multimedia data, allowing the user to quickly browse the main content of the multimedia data based on the target playback time range.

Referring to Figure 8, which is a schematic diagram of a scenario for processing multimedia data according to an embodiment of this application, the multimedia data in Figure 8 is video data, and the title information of the media data is "The Mystery of Dinosaur Extinction". It can be seen that the multimedia data mainly plays content related to the extinction of dinosaurs through video footage. Specifically, assuming that after determining the target playback time area of the multimedia data in step S102, the playback prompt information generated based on the target playback time area can be used as the indicator time area in the video progress bar in Figure 8, that is, the indicator time area can serve as the playback prompt information corresponding to the multimedia data. Alternatively, at the position of the target playback time area in the video progress bar, the frame image corresponding to the target playback time area is displayed to the user as playback prompt information, that is, the video content used to explain that "dinosaurs have become extinct" is used as playback prompt information, and the user is prompted with the playback time area of the main content related to the title information "The Mystery of Dinosaur Extinction".

Optionally, when multiple target playback time zones are identified, multiple playback prompts can be generated to indicate the playback time zones of multiple contents associated with the title information to the user. For example, when the multimedia data is movie data, multiple playback prompts can be generated based on multiple target playback time zones to indicate the playback time zones of multiple main contents (such as the climax) in the movie, which helps to improve the user's viewing experience.

Optionally, if the target playback time range of the multimedia data cannot be determined, it can be determined that the playback content of the multimedia data may have low relevance to the title information, meaning the playback content of the multimedia data may lack substantive content. In this case, content prompts can be generated and displayed to the user while the multimedia data is playing, informing the user that the currently playing multimedia data may not contain content related to the title information, thereby reducing the time wasted by the user on the multimedia data and making it more applicable.

In some feasible implementations, since the aforementioned playback prompt information corresponds to the main content of the multimedia data, the multimedia data is filtered based on the target playback time range. For example, multimedia data whose target playback time range duration exceeds a certain duration threshold is identified as target multimedia data. That is, the playback duration corresponding to the main content of the target multimedia data occupies a large proportion of the total playback duration of the multimedia data, which indicates that the target multimedia data contains relatively little content unrelated to the title information.

Optionally, the number of target playback time ranges can also be used as a filtering criterion to select target multimedia data with a large number of target playback time ranges. That is, the target multimedia data at this time contains multiple segments of content related to the title information.

The above method of filtering multimedia data based on the target playback time range is only an example. The specific method can be determined based on the actual application scenario requirements, and no restrictions are imposed here.

For short video apps, filtering short videos based on target playback time ranges allows for better recommendation and management. For example, apps can review and filter user-uploaded videos based on their target playback time range, rejecting those with shorter durations within that range, thus improving the overall video quality. Alternatively, when recommending videos to users, apps can prioritize videos with longer or more target playback time ranges to enhance the user's viewing experience.

Optionally, the recommended measurement for multimedia data can be determined based on the target playback time range of the multimedia data. This could be based on the user's playback habits and the target playback time range of the multimedia data. The playback habits information includes, but is not limited to, the duration of historical multimedia data played by the user (i.e., the time required to play each historical multimedia data piece), and the playback time of the user's corresponding historical multimedia data (the time consumed by the user browsing and/or listening to each historical multimedia data piece), etc. The specific details can be determined based on the actual application scenario requirements and are not limited here.

Furthermore, for short video applications, if it is determined based on a user's historical short video viewing history that the user prefers to browse shorter short videos, or that the user spends relatively little time browsing each historical short video, it indicates that the user has limited tolerance for short video content unrelated to the title information. Therefore, short video applications can recommend short videos that are earlier in the target playback time range to the user, so that the user can browse video content related to the title information in a shorter time.

In some feasible implementations, when generating playback prompt information based on the target playback time range of multimedia data, if the playback prompt information is associated with the text information corresponding to the target playback time range, a corresponding multimedia data recommendation strategy can also be determined based on the playback prompt information, such as recommending multimedia data related to the multimedia data that the user has played in the past.

The playback prompt information is associated with the text information corresponding to the target playback time area. It may be the frame image or text information corresponding to the target playback time area, or the keywords or category tags corresponding to the frame image or text information. The specific information can be determined based on the actual application scenario requirements and is not limited here.

For example, for short video applications, if it can be determined from the user's historical short video playback data that the user frequently plays short videos related to "football", then the short video music can identify target short videos that are associated with "football" in the playback prompt information and recommend target short videos to the user.

Furthermore, after identifying target short videos whose playback prompts are associated with "football," users can further filter suitable target short videos based on the target playback time range corresponding to the playback prompts of each target short video. For example, target short videos with earlier playback time ranges can be selected and recommended to users first.

For example, music apps can determine the types of music a user frequently plays by analyzing their historical music playback data, such as "inspirational music," "love songs," and "English songs." Based on the playback prompts for each song, the app can then recommend related music to the user, thereby increasing user engagement.

Optionally, if the playback prompt information is associated with text information corresponding to the target playback time area, a corresponding multimedia data management strategy can also be determined based on the playback prompt information. For example, multimedia data can be classified based on the playback prompt information corresponding to each multimedia data, or after classifying the multimedia data based on the title information of each multimedia data, the multimedia data under each category can be further classified based on the playback prompt information corresponding to each multimedia data, or multimedia data in each category that does not match the category can be identified based on the playback prompt information. The specific management strategy can be determined based on the actual application scenario requirements and is not limited here.

In this embodiment, by representing multimedia data with at least one text message, the matching degree between the title information and each text message can be accurately determined at the text dimension. The matching degree is determined by the text similarity between the title information and each text message, and the frequency of each keyword appearing in each text message. This provides multiple ways to measure the correlation between title information and text messages, and thus offers multiple methods for determining the target playback time zone in multimedia data, better suited to different application scenarios. Furthermore, by combining different methods for determining the target playback time zone, the methods for determining the target playback time can be further expanded, and the risk of failure due to determining the target playback time zone based on a single method can be reduced. Moreover, processing multimedia data using the target playback time zone allows users to quickly determine the playback time zone of multimedia content related to the title information. Multimedia reminders can save users browsing time on multimedia data that does not contain multimedia content related to the title information, enhancing user appeal and demonstrating high applicability.

Referring to Figure 9, Figure 9 is a schematic diagram of the structure of a multimedia data processing apparatus provided in an embodiment of this application. The processing apparatus 1 provided in this embodiment includes:

The acquisition unit 11 is used to acquire at least one text information contained in the multimedia data, as well as the title information of the multimedia data.

Determining unit 12 is used to determine the matching degree between the above title information and each of the above text information;

The aforementioned determining unit 12 is used to determine the target playback time region in the aforementioned multimedia data based on the matching degree corresponding to each of the aforementioned text information.

The playback unit 13 is used to process the multimedia data according to the target playback time range.

In some feasible implementations, the determining unit 12 is used for:

Determine the text similarity between the above title information and each of the above text information, and use the above text similarity as the matching score;

Identify the keywords in the title information above. For each piece of text information above, determine the matching degree between the title information and the text information based on the number of times each of the above keywords appears in the text information.

In some feasible implementations, the determining unit 12 is used for:

Determine the weight of each of the above keywords;

For each of the above text information, the matching degree between the title information and the text information is determined based on the number of times each of the above keywords appears in the text information and the weight of each of the above keywords.

In some feasible implementations, the determining unit 12 is used for:

The playback time area corresponding to the text information that meets the matching conditions is determined as the target playback time area in the above multimedia data.

The matching criteria above include any of the following:

The matching degree is higher than the matching degree threshold;

The match score is the highest possible.

In some feasible implementations, if the text information that satisfies the above matching conditions includes at least two text information pieces, the determining unit 12 is used to:

Based on the playback time of each of the above-mentioned text messages that meet the above-mentioned matching conditions in the above-mentioned multimedia data, the playback time area of the first text message that meets the above-mentioned matching conditions is determined as the target playback time area in the above-mentioned multimedia data.

In some feasible implementations, the determining unit 12 is used for:

If none of the above text information contains the specified information, then the text similarity between the title information and each of the above text information is determined. If there is a text similarity that meets the preset conditions, then the text similarity corresponding to each of the above text information is used as the matching degree.

If there is no text similarity among the text similarities corresponding to each of the above text information that meets the above preset conditions, then the matching degree between the title information and each of the above text information is determined based on the number of times each of the above keywords appears in each of the above text information.

In some feasible implementations, the determining unit 12 is further configured to:

The playback time range corresponding to the text information containing the specified information in each of the above text information is determined as the target playback time range in the above multimedia data.

In some feasible implementations, the playback unit 13 described above is used for:

Upon receiving a playback request for the aforementioned multimedia data, playback prompt information corresponding to the aforementioned multimedia data is generated based on the aforementioned target playback time range, wherein the aforementioned playback prompt information is used to indicate the aforementioned target playback time range;

Play the aforementioned multimedia data and display the aforementioned playback prompt information to the user.

In some feasible implementations, the multimedia data is video data; the acquisition unit 11 is used for:

Obtain the caption information of at least one frame of the above video data, and use the caption information of the at least one frame of the above video data as at least one piece of text information contained in the above video data.

The caption information for one frame of an image is a text message.

In some feasible implementations, the multimedia data is audio data; the acquisition unit 11 is used for:

Speech recognition is performed on the above audio data to obtain the speech recognition result of the above audio data;

The text content corresponding to at least one statement in the above speech recognition results shall be used as at least one piece of text information contained in the above audio data.

In practice, the device 1 can execute the implementation methods provided by the steps in Figure 1 above through its built-in functional modules. For details, please refer to the implementation methods provided by the steps above, which will not be repeated here.

In this embodiment, by representing multimedia data with at least one text message, the matching degree between the title information and each text message can be accurately determined at the text dimension. The matching degree is determined by the text similarity between the title information and each text message, and the frequency of each keyword appearing in each text message. This provides multiple ways to measure the correlation between title information and text messages, and thus offers multiple methods for determining the target playback time zone in multimedia data, better suited to different application scenarios. Furthermore, by combining different methods for determining the target playback time zone, the methods for determining the target playback time can be further expanded, and the risk of failure due to determining the target playback time zone based on a single method can be reduced. Moreover, processing multimedia data using the target playback time zone allows users to quickly determine the playback time zone of multimedia content related to the title information. Multimedia reminders can save users browsing time on multimedia data that does not contain multimedia content related to the title information, enhancing user appeal and demonstrating high applicability.

Referring to Figure 10, which is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, the electronic device 1000 in this embodiment may include a processor 1001, a network interface 1004, and a memory 1005. Furthermore, the electronic device 1000 may also include a user interface 1003 and at least one communication bus 1002. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen and a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1004 may be a high-speed RAM or a non-volatile memory, such as at least one disk storage device. The memory 1005 may optionally be at least one storage device located remotely from the processor 1001. As shown in Figure 10, the memory 1005, as a computer-readable storage medium, may include an operating system, a network communication module, a user interface module, and a device control application.

In the electronic device 1000 shown in Figure 10, the network interface 1004 provides network communication functionality; the user interface 1003 is mainly used to provide an input interface for the user; and the processor 1001 can be used to call the device control application stored in the memory 1005 to achieve:

Obtain at least one text information contained in the multimedia data, as well as the title information of the multimedia data;

Determine the degree of matching between the above title information and each of the above text information;

Based on the matching degree of each of the above text information, the target playback time area in the above multimedia data is determined;

The multimedia data is processed according to the target playback time range.

In some feasible implementations, the processor 1001 described above is used for:

Determine the text similarity between the above title information and each of the above text information, and use the above text similarity as the matching score;

Identify the keywords in the title information above. For each piece of text information above, determine the matching degree between the title information and the text information based on the number of times each of the above keywords appears in the text information.

In some feasible implementations, the processor 1001 described above is used for:

Determine the weight of each of the above keywords;

For each of the above text information, the matching degree between the title information and the text information is determined based on the number of times each of the above keywords appears in the text information and the weight of each of the above keywords.

In some feasible implementations, the processor 1001 described above is used for:

The playback time area corresponding to the text information that meets the matching conditions is determined as the target playback time area in the above multimedia data.

The matching criteria above include any of the following:

The matching degree is higher than the matching degree threshold;

The match score is the highest possible.

In some feasible implementations, if the text information that satisfies the above matching conditions includes at least two text information pieces, the processor 1001 is used to:

Based on the playback time of each of the above-mentioned text messages that meet the above-mentioned matching conditions in the above-mentioned multimedia data, the playback time area of the first text message that meets the above-mentioned matching conditions is determined as the target playback time area in the above-mentioned multimedia data.

In some feasible implementations, the processor 1001 described above is used for:

If none of the above text information contains the specified information, then the text similarity between the title information and each of the above text information is determined. If there is a text similarity that meets the preset conditions, then the text similarity corresponding to each of the above text information is used as the matching degree.

If there is no text similarity among the text similarities corresponding to each of the above text information that meets the above preset conditions, then the matching degree between the title information and each of the above text information is determined based on the number of times each of the above keywords appears in each of the above text information.

In some feasible implementations, the processor 1001 is further configured to:

The playback time range corresponding to the text information containing the specified information in each of the above text information is determined as the target playback time range in the above multimedia data.

In some feasible implementations, the processor 1001 described above is used for:

Upon receiving a playback request for the aforementioned multimedia data, playback prompt information corresponding to the aforementioned multimedia data is generated based on the aforementioned target playback time range, wherein the aforementioned playback prompt information is used to indicate the aforementioned target playback time range;

Play the aforementioned multimedia data and display the aforementioned playback prompt information to the user.

In some feasible implementations, the multimedia data is video data; the processor 1001 is used for:

Obtain the caption information of at least one frame of the above video data, and use the caption information of the at least one frame of the above video data as at least one piece of text information contained in the above video data.

The caption information for one frame of an image is a text message.

In some feasible implementations, the multimedia data is audio data; the processor 1001 is used for:

Speech recognition is performed on the above audio data to obtain the speech recognition result of the above audio data;

The text content corresponding to at least one statement in the above speech recognition results shall be used as at least one piece of text information contained in the above audio data.

It should be understood that in some feasible implementations, the processor 1001 described above may be a central processing unit (CPU), which may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor. The memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of the memory may also include non-volatile random access memory. For example, the memory may also store device type information.

In practice, the aforementioned electronic device 1000 can execute the implementation methods provided in the steps shown in Figure 1 above through its built-in functional modules. For details, please refer to the implementation methods provided in the steps above, which will not be repeated here.

In this embodiment, by representing multimedia data with at least one text message, the matching degree between the title information and each text message can be accurately determined at the text dimension. The matching degree is determined by the text similarity between the title information and each text message, and the frequency of each keyword appearing in each text message. This provides multiple ways to measure the correlation between title information and text messages, and thus offers multiple methods for determining the target playback time zone in multimedia data, better suited to different application scenarios. Furthermore, by combining different methods for determining the target playback time zone, the methods for determining the target playback time can be further expanded, and the risk of failure due to determining the target playback time zone based on a single method can be reduced. Moreover, processing multimedia data using the target playback time zone allows users to quickly determine the playback time zone of multimedia content related to the title information. Multimedia reminders can save users browsing time on multimedia data that does not contain multimedia content related to the title information, enhancing user appeal and demonstrating high applicability.

This application also provides a computer-readable storage medium storing a computer program that is executed by a processor to implement the methods provided in the steps of FIG1 above. For details, please refer to the implementation methods provided in the steps above, which will not be repeated here.

The aforementioned computer-readable storage medium can be an internal storage unit of the apparatus or device provided in any of the foregoing embodiments, such as a hard disk or memory of an electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device. The aforementioned computer-readable storage medium can also include magnetic disks, optical disks, read-only memory (ROM), or random access memory (RAM), etc. Furthermore, the computer-readable storage medium can include both internal storage units and external storage devices of the electronic device. The computer-readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

This application provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the steps of FIG1.

The terms "first," "second," etc., used in the claims, description, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or electronic device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or electronic devices. References to "embodiment" herein mean that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The presentation of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments. The term "and/or" as used in this application's description and appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Those skilled in the art can implement the described functions using different methods for each specific application, but such implementations should not be considered beyond the scope of this application.

The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims (8)

1. A method for processing multimedia data, characterized in that the method comprises: Obtain at least two text information items contained in the multimedia data, as well as the title information of the multimedia data; Determine the matching degree between the title information and each of the text information; If the matching degree between the title information and the first text information is higher than the matching degree threshold, and the time distance between the playback time area corresponding to the first text information and the playback time area corresponding to the second text information is not less than the time distance threshold, then the playback time area corresponding to the first text information is determined as the target playback time area in the multimedia data; the first text information and the second text information are two different text information among the at least two text information, and the playback time area corresponding to the second text information is before the playback time area corresponding to the first text information; the playback time area corresponding to the second text information is the target playback time area in the multimedia data; Upon receiving a playback request for the multimedia data, playback prompt information corresponding to the multimedia data is generated based on the target playback time region, wherein the playback prompt information includes the frame image corresponding to the target playback time region; Play the multimedia data and display the playback prompt information to the user. 2. The method according to claim 1, wherein determining the matching degree between the title information and each of the text information comprises: If none of the text information contains the specified information, then the text similarity between the title information and each of the text information is determined. If there is a text similarity that meets the preset conditions, then the text similarity corresponding to each of the text information is used as the matching degree. If there is no text similarity among the text similarities corresponding to each of the text information that meets the preset conditions, then the matching degree between the title information and each of the text information is determined based on the number of times each keyword of the title information appears in each of the text information. 3. The method according to claim 1 or 2, characterized in that the method further comprises: The playback time region corresponding to the text information containing the specified information in each of the aforementioned text information is determined as the target playback time region in the multimedia data. 4. The method according to claim 1, wherein the multimedia data is video data; and the acquisition of at least two text information items contained in the multimedia data includes: Obtain caption information from at least two frames of the video data, and use the caption information from the at least two frames of the video data as at least two pieces of text information contained in the video data; The caption information for one frame of an image is a text message. 5. The method according to claim 1, wherein the multimedia data is audio data; and the acquisition of at least two text information items contained in the multimedia data includes: Perform speech recognition on the audio data to obtain the speech recognition result of the audio data; The text content corresponding to at least two sentences in the speech recognition result shall be used as at least two text information contained in the audio data. 6. A multimedia data processing apparatus, characterized in that the processing apparatus comprises: The acquisition unit is used to acquire at least two text information items contained in the multimedia data, as well as the title information of the multimedia data; A determining unit is used to determine the matching degree between the title information and each of the text information; The determining unit is configured to determine the playback time region corresponding to the first text information as the target playback time region in the multimedia data if the matching degree between the title information and the first text information is higher than the matching degree threshold, and the time distance between the playback time region corresponding to the first text information and the playback time region corresponding to the second text information is not less than the time distance threshold; the first text information and the second text information are two different text information among the at least two text information, and the playback time region corresponding to the second text information is before the playback time region corresponding to the first text information; the playback time region corresponding to the second text information is the target playback time region in the multimedia data. A playback unit is configured to generate playback prompt information corresponding to the multimedia data based on the target playback time region when a playback request for the multimedia data is received, wherein the playback prompt information includes the frame image corresponding to the target playback time region; Play the multimedia data and display the playback prompt information to the user. 7. An electronic device, characterized in that it comprises a processor and a memory, the processor and the memory being interconnected; The memory is used to store computer programs; The processor is configured to perform the method as described in any one of claims 1 to 5 when the computer program is invoked. 8. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, the computer program being executed by a processor to implement the method of any one of claims 1 to 5.