CN114564614A - A kind of video clip automatic search method, system, device and readable storage medium - Google Patents

Abstract

The invention provides a method, a system, a device and a readable storage medium for automatically searching video clips, wherein the method comprises the following steps: uploading the video streaming to a video storage device by a video creator, and synchronously establishing a fragment keyword database of the streaming video through a voice recognition AI model; respectively associating a video search engine with a video storage device and a keyword database; a watching user logs in a video search engine through a client and inputs keywords, and the video search engine finds corresponding video segments from a video storage device through the keywords and pushes the video segments to the client; and the watching user plays the pushed video clip through the client. The invention records the video content by automatically recognizing the voice and the semantics and realizes the automatic video segment recording and searching with low operation resources.

Description

A kind of video clip automatic search method, system, device and readable storage medium

technical field

The present invention relates to the technical field of video data search, and more particularly to an automatic search method, system, device and readable storage medium for video clips.

Background technique

With the continuous development of multimedia services, the number of videos is increasing. In order to get rid of the traditional way of passively watching TV videos, more and more people choose to search for the videos they want to watch through the Internet. Therefore, video search has also become a key link in obtaining videos.

At present, the video search function is usually implemented by means of keyword search. This method can only search for videos that match the title or description, and you need to watch the video to know the minutes and seconds of the desired segment, or you need to watch the video to know whether you have the desired content. However, when you want to quickly find information, it is time-consuming to watch the video, so how to quickly search for interesting clips is an important issue.

For the realization of the video clip search function, the following three methods are mainly adopted in the prior art;

1. The video creator can set the paragraph by marking the time stamp in the video description, as well as the key description, so that the viewing user can directly click on the paragraph.

2. The clips are manually set in advance for subsequent viewers to click.

3. By analyzing the characteristics of video image content, and recording the characteristics and time points for subsequent searches to find clips.

In the above-mentioned existing method, the method of labeling paragraphs by the video creator requires the creator to manually label the video, and the video can only be searched through the title first, and then the watching user clicks the video before clicking the labeling paragraph; if If there is a discrepancy between the title and paragraph content, the video cannot be searched. However, the method of labeling by analyzing the content features of video images requires large computing resources due to the complicated selection and comparison of image features.

SUMMARY OF THE INVENTION

In view of the above problems, the purpose of the present invention is to provide an automatic search method, system, device and readable storage medium for video clips, which can record video content by automatically recognizing voice and semantics, and realize automatic and low computing resource video. Fragment recording and searching.

In order to achieve the above object, the present invention is realized through the following technical solutions: an automatic search method for video clips, comprising:

The video creator uploads the video stream to the video storage device, and simultaneously establishes the segment keyword database of the stream video through the voice recognition AI model;

associating the video search engine with the video storage device and the keyword database respectively;

The viewing user logs in to the video search engine through the client and inputs a keyword, and the video search engine finds the corresponding video clip from the video storage device through the keyword and pushes it to the client;

Watch users play the pushed video clips through the client.

Further, the establishment of a segment keyword database of the streaming video through the voice recognition AI model includes: performing voice recognition on the uploaded video stream through the voice recognition AI model to establish a subtitle file with a timeline;

The keywords in the subtitle file are automatically screened out as the search keywords of the video clips through the preset semantic recognition system;

The semantic recognition system builds a segment keyword database of the streaming video according to the search keywords.

Further, the semantic recognition system establishes a segment keyword database of the streaming video according to the search keywords, including:

The semantic recognition system generates corresponding video numbers and timestamps according to the search keywords;

Store search keywords, video numbers and timestamps to the clip keyword database.

Further, the video search engine finds the corresponding video clips from the video storage device through keywords and pushes them to the client, including:

The video search engine finds the same search keyword in the segment keyword database by keyword;

Obtain the corresponding video number and timestamp according to the found search keywords;

Find the corresponding video segment in the video storage device according to the video number and time stamp;

The video storage device sends the video stream corresponding to the video clip to the client.

Further, the viewing user logs into the video search engine through the client and inputs keywords, specifically:

Watch users log in to the video search engine through the client and enter advanced keywords through the AND operator.

Correspondingly, the present invention also discloses an automatic search system for video clips, including:

The video stream input module is used for video creators to upload video streams to the video storage device, and synchronously establish the segment keyword database of the stream video through the voice recognition AI model;

The association module is used to associate the video search engine with the video storage device and the keyword database respectively; the search module is used to watch the user log in to the video search engine through the client and input the keyword, and the video search engine uses the keyword from the video storage device. Find the corresponding video clip in the client and push it to the client; the playback module is used to watch the user play the pushed video clip through the client.

Further, the video stream input module includes:

The font file building unit is used to perform speech recognition on the uploaded video stream through the speech recognition AI model to create a subtitle file with a timeline;

A keyword screening unit for automatically screening out keywords in the subtitle file as search keywords for video clips through a preset semantic recognition system;

The segment keyword database building unit is used to establish a segment keyword database of the streaming video according to the search keyword through the semantic recognition system.

Further, the search module includes:

The comparison unit is used for the video search engine to find the same search keyword in the segment keyword database through keywords;

Mark acquisition unit, for obtaining corresponding video number and time stamp according to the search keyword found; Video segment acquisition unit, for finding corresponding video segment in the video storage device according to video number and time stamp;

The video stream push unit is used for sending the video stream corresponding to the video clip to the client through the video storage device.

Correspondingly, the present invention discloses a video clip automatic search device, comprising:

Memory, used to store the automatic search program for video clips;

The processor is configured to implement the steps of the automatic video clip search method according to any one of the above when executing the video clip automatic search program.

Correspondingly, the present invention discloses a readable storage medium, on which an automatic video segment search program is stored, and when the video segment automatic search program is executed by a processor, the video as described in any of the above is realized. Steps for the Fragment AutoSearch method.

Compared with the prior art, the beneficial effects of the present invention are:

1. The present invention provides a video clip automatic search method, system, device and readable storage medium. Through the voice recognition AI model and the semantic recognition system, a fragment keyword database of streaming video is automatically established, and keywords can be used. Search the clip keyword database to get the corresponding video clip by video number and timestamp.

2. The present invention can automatically create key segments without manual annotation.

3. The computing resources required for the intelligent speech/semantic recognition algorithm in the present invention are far less than the computing resources required for the image recognition algorithm.

4. Since the computing resources of the intelligent speech/semantic recognition algorithm are small, the present invention can recognize videos with high resolution and high frame rate in real time.

It can be seen that, compared with the prior art, the present invention has outstanding substantive features and significant progress, and the beneficial effects of its implementation are also obvious.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the provided drawings without creative efforts.

FIG. 1 is a flow chart of a method of a specific embodiment of the present invention.

FIG. 2 is a system structure diagram of a specific embodiment of the present invention.

In the figure, 1 is a video stream input module; 2 is an association module; 3 is a search module; 4 is a playback module; 11 is a font file building unit; 12 is a keyword screening unit; 13 is a segment keyword database building unit; 31 is a comparison unit; 32 is a mark acquisition unit; 33 is a video clip acquisition unit; 34 is a video stream push unit.

Detailed ways

The core of the present invention is to provide an automatic search method for video clips. In the prior art, the method of labeling paragraphs by video creators requires manual labeling by the creators, and videos can only be searched through the title first, and then the viewing user clicks After the video, you can click and mark the paragraph again; if there is a discrepancy between the title and the content of the paragraph, the video cannot be searched. However, the method of labeling by analyzing the content features of video images requires large computing resources due to the complicated selection and comparison of image features.

In the automatic search method for video clips provided by the present invention, firstly, the video creator uploads the video stream to the video storage device, and simultaneously establishes a clip keyword database of the stream video through the voice recognition AI model. Then, the video search engine is associated with the video storage device and the keyword database, respectively. At this point, the viewing user logs in to the video search engine through the client and inputs a keyword, and the video search engine finds the corresponding video clip from the video storage device through the keyword and pushes it to the client; finally, the viewing user plays the pushed video through the client Fragment. It can be seen that the present invention records video content by automatically recognizing voice and semantics, and realizes automatic recording and searching of video clips with low computing resources.

In order to make those skilled in the art better understand the solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are only some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

Example 1:

As shown in FIG. 1, this embodiment provides an automatic search method for video clips, including the following steps:

S1: The video creator uploads the video stream to the video storage device, and simultaneously establishes a segment keyword database of the stream video through the voice recognition AI model.

Among them, when building the segment keyword database of the streaming video, firstly, the speech recognition AI model is used to perform speech recognition on the uploaded video stream to create a subtitle file with a timeline. Then, keywords in the subtitle file are automatically screened out as search keywords for video clips through a preset semantic recognition system. Finally, the semantic recognition system builds a segment keyword database of the streaming video according to the search keywords.

As an example, when using a semantic recognition system to create a segment keyword database of streaming videos based on search keywords, it is first necessary to perform speech recognition on the uploaded video stream through a speech recognition AI model to create a subtitle file with a timeline. Then, keywords in the subtitle file are automatically screened out as search keywords for video clips through a preset semantic recognition system. Finally, a segment keyword database of the streaming video is established by the semantic recognition system according to the search keywords.

This step is based on the premise that the current speech recognition AI model is quite mature and accurate, and has been widely used in the application of automatic subtitles. Therefore, when a video creator uploads a video stream, speech recognition can be performed at the same time to automatically create subtitles with a timeline.

When running, through the semantic recognition system, the keyword in the subtitle file can be automatically selected as the search keyword for the video clip, and the video number and the clip timeline (in the format of XX:XX:XX,XXX) are established for this keyword results. The speech recognition AI model and the semantic recognition system can be integrated into a single "keyword automatic recognition" engine, and through this engine, a segment keyword database of streaming video can be created in real time.

S2: Associate the video search engine with the video storage device and the keyword database respectively.

S3: The viewing user logs in to the video search engine through the client and inputs a keyword, and the video search engine finds the corresponding video clip from the video storage device through the keyword and pushes it to the client.

Specifically, the video search engine first finds the same search keyword in the segment keyword database through keywords. At this time, the corresponding video number and time stamp are obtained according to the found search keyword. Then, the corresponding video segment is found in the video storage device according to the video number and time stamp. Finally, the video storage device sends the video stream corresponding to the video clip to the client.

In addition, when entering keywords, advanced keywords can be entered through the AND operator. for advanced search.

As an example, the video clip search process is as follows:

1. The user enters a keyword.

2. Keyword search through video search engine.

3. Get video numbers and timestamps that match the search criteria in the segment keyword database.

4. Find the corresponding video clip from the video storage device.

5. Recommend video clips to users.

S4: Watch the video clips pushed by the user through the client.

This embodiment provides an automatic search method for video clips. Through the AI model of speech recognition and the semantic recognition system, a clip keyword database of streaming video is automatically established, and keywords can be used to search the clip keyword database, and the video number can be used to search the clip keyword database. and timestamp to get the corresponding video clip. The method records the video content by automatically recognizing voice and semantics, and realizes the automatic recording and searching of video clips with low computing resources.

Embodiment 2:

Based on the first embodiment, as shown in FIG. 2 , the present invention also discloses an automatic video clip search system, including: a video stream input module 1 , an association module 2 , a search module 3 and a playback module 4 .

The video stream input module 1 is used for the video creator to upload the video stream to the video storage device, and synchronously establish the segment keyword database of the stream video through the voice recognition AI model.

Specifically, the video stream input 1 module includes: a font file building unit 11 , a keyword screening unit 12 and a segment keyword database building unit 13 .

The font file building unit 11 is used to perform speech recognition on the uploaded video stream through the speech recognition AI model, so as to create a subtitle file with a timeline.

The keyword screening unit 12 is configured to automatically filter out keywords in the subtitle file as search keywords for video clips through a preset semantic recognition system.

The segment keyword database building unit 13 is configured to establish a segment keyword database of the streaming video according to the search keyword through the semantic recognition system.

The association module 2 is used to associate the video search engine with the video storage device and the keyword database respectively.

The search module 3 is used to watch the user log in to the video search engine through the client and input a keyword, and the video search engine finds the corresponding video clip from the video storage device through the keyword and pushes it to the client.

Specifically, the search module 3 includes: a comparison unit 31 , a marker acquisition unit 32 , a video segment acquisition unit 33 and a video stream push unit 34 .

The comparison unit 31 is used for the video search engine to find the same search keyword in the segment keyword database through keywords.

The tag obtaining unit 32 is configured to obtain the corresponding video number and time stamp according to the found search keyword.

The video clip obtaining unit 33 is configured to find the corresponding video clip in the video storage device according to the video number and the time stamp.

The video stream pushing unit 34 is configured to send the video stream corresponding to the video clip to the client through the video storage device.

The playing module 4 is used to watch the video clips that the user plays and pushes through the client.

This embodiment provides an automatic search system for video clips, which can automatically establish a clip keyword database of streaming videos through a voice recognition AI model and a semantic recognition system, and use keywords to search the clip keyword database. and timestamp to get the corresponding video clip.

Embodiment three:

The present embodiment discloses a video clip automatic search device, including a processor and a memory; wherein, the processor implements the following steps when executing the video clip automatic search program saved in the memory:

1. The video creator uploads the video stream to the video storage device, and simultaneously establishes the segment keyword database of the stream video through the voice recognition AI model.

2. Associate the video search engine with the video storage device and the keyword database respectively.

3. The viewing user logs in to the video search engine through the client and inputs a keyword, and the video search engine finds the corresponding video clip from the video storage device through the keyword and pushes it to the client.

4. Watch the video clips that users play through the client.

Further, the apparatus for automatically searching for video clips in this embodiment may further include:

The input interface is used to obtain the video clip automatic search program imported from the outside world, and save the obtained video clip automatic search program into the memory. It can also be used to obtain various instructions and parameters transmitted by the external terminal equipment, and transmit the into the processor, so that the processor can use the above various instructions and parameters to carry out corresponding processing. In this embodiment, the input interface may specifically include, but is not limited to, a USB interface, a serial interface, a voice input interface, a fingerprint input interface, a hard disk reading interface, and the like.

The output interface is used to output various data generated by the processor to the terminal equipment connected to it, so that other terminal equipment connected to the output interface can obtain various data generated by the processor. In this embodiment, the output interface may specifically include, but is not limited to, a USB interface, a serial interface, and the like.

The communication unit is used for establishing a long-distance communication connection between the video clip automatic searching device and the external server, so that the video clip automatic searching device can mount the image file to the external server. In this embodiment, the communication unit may specifically include, but is not limited to, a remote communication unit based on a wireless communication technology or a wired communication technology.

The keyboard is used to obtain various parameter data or instructions input by the user by hitting the keycap in real time.

The display is used to display the relevant information of the short-circuit positioning process of the power supply line of the server in real time.

A mouse can be used to assist the user in entering data and simplify the user's operations.

Embodiment 4:

This embodiment also discloses a readable storage medium, where the readable storage medium includes random access memory (RAM), internal memory, read only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, Register, hard disk, removable hard disk, CD-ROM, or any other form of storage medium known in the art. The readable storage medium stores a video clip automatic search program, and when the video clip automatic search program is executed by the processor, the following steps are implemented:

1. The video creator uploads the video stream to the video storage device, and simultaneously establishes the segment keyword database of the stream video through the voice recognition AI model.

2. Associate the video search engine with the video storage device and the keyword database respectively.

3. The viewing user logs in to the video search engine through the client and inputs a keyword, and the video search engine finds the corresponding video clip from the video storage device through the keyword and pushes it to the client.

4. Watch the video clips that users play through the client.

To sum up, the present invention records video content by automatically recognizing voice and semantics, and realizes automatic recording and searching of video clips with low computing resources.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the various embodiments may be referred to each other. For the method disclosed in the embodiment, since it corresponds to the system disclosed in the embodiment, the description is relatively simple, and the relevant part can be referred to the description of the method.

Professionals may further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two, in order to clearly illustrate the possibilities of hardware and software. Interchangeability, the above description has generally described the components and steps of each example in terms of functionality. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of the present invention.

In the several embodiments provided by the present invention, it should be understood that the disclosed systems, systems and methods may be implemented in other manners. For example, the system embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of systems or units, and may be in electrical, mechanical or other forms.

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

In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, or each module may exist physically alone, or two or more modules may be integrated into one unit.

Similarly, each processing unit in each embodiment of the present invention may be integrated into one functional module, or each processing unit may exist physically, or two or more processing units may be integrated into one functional module.

The steps of a method or algorithm described in conjunction with the embodiments disclosed herein may be directly implemented in hardware, a software module executed by a processor, or a combination of the two. A software module can be placed in random access memory (RAM), internal memory, read only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other in the art. in any other known form of storage medium.

Finally, it should also be noted that in this document, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply these entities or there is any such actual relationship or sequence between operations. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device comprising a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The automatic search method, system, device and readable storage medium for video clips provided by the present invention have been described in detail above. The principles and implementations of the present invention are described herein by using specific examples, and the descriptions of the above embodiments are only used to help understand the method and the core idea of the present invention. It should be pointed out that for those skilled in the art, without departing from the principle of the present invention, several improvements and modifications can also be made to the present invention, and these improvements and modifications also fall within the protection scope of the claims of the present invention.

Claims (10)

1. An automatic video segment searching method, comprising:

uploading the video streaming to a video storage device by a video creator, and synchronously establishing a fragment keyword database of the streaming video through an AI model;

respectively associating a video search engine with a video storage device and a keyword database;

a watching user logs in a video search engine through a client and inputs keywords, and the video search engine finds corresponding video segments from a video storage device through the keywords and pushes the video segments to the client;

and the watching user plays the pushed video clip through the client.

2. The method of claim 1, wherein the creating a segment keyword library of the streaming video through the AI model comprises:

performing voice recognition on the uploaded video streaming through a voice recognition AI model to establish a subtitle file with a time axis;

automatically screening out keywords in the subtitle files as search keywords of the video clips through a preset semantic recognition system;

the semantic recognition system establishes a fragment keyword database of the streaming video according to the search keywords.

3. The method of claim 2, wherein the semantic recognition system builds a segment keyword database of the streaming video according to the search keywords, comprising:

the semantic recognition system generates corresponding video numbers and time stamps according to the search keywords;

the search key, the video number and the time stamp are stored to the segment key db.

4. The method of claim 3, wherein the video search engine finds a corresponding video segment from a video storage device by using a keyword and pushes the video segment to the client, and comprises:

The video search engine finds the same search keyword in the fragment keyword database through the keyword;

obtaining corresponding video numbers and time stamps according to the found search keywords;

finding out a corresponding video clip in a video storage device according to the video number and the time stamp;

and the video storage device sends the video streaming corresponding to the video clip to the client.

5. The method according to claim 4, wherein the viewing user logs in the video search engine through the client and inputs keywords, specifically:

the watching user logs in the video search engine through the client AND inputs the advanced keyword through the AND operator.

6. An automatic video segment searching system, characterized in that,

the video streaming input module is used for uploading video streaming to the video storage device by a video creator and synchronously establishing a fragment keyword database of the streaming video through a voice recognition AI model;

the correlation module is used for correlating the video search engine with the video storage device and the keyword database respectively;

the search module is used for enabling a watching user to log in a video search engine through a client and input keywords, and the video search engine finds corresponding video segments from the video storage device through the keywords and pushes the video segments to the client;

And the playing module is used for watching the video clips pushed by the user through the client.

7. The system of claim 6, wherein the video stream input module comprises:

the word stock file building unit is used for carrying out voice recognition on the uploaded video streaming through a voice recognition AI model so as to build a subtitle file with a time axis;

the keyword screening unit is used for automatically screening out keywords in the subtitle files as search keywords of the video clips through a preset semantic recognition system;

and the fragment keyword database building unit is used for building a fragment keyword database of the streaming video according to the search keywords through the semantic recognition system.

8. The video clip auto-search system of claim 7, wherein the search module comprises:

the comparison unit is used for finding the same search key words in the fragment key word database by the video search engine through the key words;

a mark acquisition unit for acquiring a corresponding video number and a time stamp according to the found search keyword;

a video clip acquiring unit for finding a corresponding video clip in the video storage device according to the video number and the time stamp;

And the video streaming pushing unit is used for sending the video streaming corresponding to the video clip to the client through the video storage device.

9. An apparatus for automatically searching for a video clip, comprising:

a memory for storing a video clip automatic search program;

a processor for implementing the steps of the video segment automatic search method according to any one of claims 1 to 5 when executing the video segment automatic search program.

10. A readable storage medium, characterized by: the readable storage medium has stored thereon a video clip automatic search program, which when executed by a processor, implements the steps of the video clip automatic search method according to any one of claims 1 to 5.

Images