Interview video uploading method and device, computer equipment and storage medium

Abstract

The invention provides an interview video uploading method and device, computer equipment and a storage medium, belongs to the technical field of AI interview, and can solve the problem that final interview submission is unsuccessful due to overlong waiting time, jamming or interruption in the existing interview video uploading process. The interview video uploading method comprises the following steps: segmenting an interview video into a plurality of video clips according to preset key contents, and extracting key clips and non-key clips from the video clips; the key segment corresponds to preset key content; and uploading the key segment and the non-key segment according to the network state.

Description

Technical Field

[0001] The present disclosure belongs to the field of AI interview technology, and specifically relates to a method, device, computer equipment and storage medium for uploading interview videos. Background Art

[0002] With the advancement of science and technology and the rapid development of artificial intelligence, traditional human resource management methods are undergoing profound changes. In particular, in the recruitment process, interviews are a crucial step in assessing a candidate's abilities and potential. Their efficiency and quality are directly related to the effectiveness of a company's human resource management.

[0003] In recent years, a number of interview systems based on AI (Artificial Intelligence) technology have emerged on the market. These AI interview systems leverage advanced algorithms to perform preliminary intelligent analysis and assessment of candidates' verbal behavior, thereby enhancing interview standardization and improving efficiency. Summary of the Invention

[0004] The present disclosure aims to solve at least one of the technical problems existing in the prior art, and provides an interview video uploading method, device, computer equipment and storage medium.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for uploading an interview video, the method comprising:

[0006] According to the preset key content, the interview video is divided into multiple video segments, and key segments and non-key segments are extracted from the video segments; the key segments correspond to the preset key content;

[0007] The key segment and the non-key segment are uploaded according to the network status.

[0008] In some embodiments, the step of segmenting the interview video into multiple video segments based on preset key content, and extracting key segments and non-key segments from the video segments, includes:

[0009] Using a speech recognition engine, recognizing speech content in the video clip and converting the speech content into text content;

[0010] Using natural language processing technology to perform semantic analysis on the text content, and using natural language processing models for different positions to identify key text in the text content that is relevant to the corresponding position;

[0011] According to the timestamp corresponding to the key text on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

[0012] In some embodiments, the step of segmenting the interview video into multiple video segments based on preset key content, and extracting key segments and non-key segments from the video segments, includes:

[0013] Use image recognition algorithms to identify the interviewee's facial expressions and body movements in each frame, and mark the key facial expressions and key body movements;

[0014] According to the timestamps corresponding to the images in which the key facial expressions and the key body movements appear on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

[0015] In some embodiments, the step of segmenting the interview video into multiple video segments based on preset key content, and extracting key segments and non-key segments from the video segments, includes:

[0016] According to the type of interview questions, pre-stored interview questions are divided into key questions and non-key questions;

[0017] According to the timestamps corresponding to the time from when the key question is asked to when the answer is completed on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

[0018] In some embodiments, the step of segmenting the interview video into multiple video segments based on preset key content, and extracting key segments and non-key segments from the video segments, includes:

[0019] According to the type of job skills, the pre-stored job skills are divided into key job skills and non-key job skills;

[0020] According to the timestamp corresponding to the interviewee's description of the key job skills on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

[0021] In some embodiments, the step of segmenting the interview video into multiple video segments based on preset key content, and extracting key segments and non-key segments from the video segments, includes:

[0022] Use eye tracking technology to collect interviewees’ eye movement data and mark key eye movement data;

[0023] According to the timestamp corresponding to the key eye movement data on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

[0024] In some embodiments, the step of segmenting the interview video into multiple video segments based on preset key content, and extracting key segments and non-key segments from the video segments, includes:

[0025] Obtain the interviewee's human-computer interaction data and mark the key human-computer interaction data;

[0026] According to the timestamp corresponding to the key human-computer interaction data on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

[0027] In some embodiments, the step of segmenting the interview video into multiple video segments based on preset key content, and extracting key segments and non-key segments from the video segments, includes:

[0028] Obtain interviewees' stress response data and mark key stress response data;

[0029] According to the timestamp corresponding to the key stress response data on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

[0030] In some embodiments, the step of segmenting the interview video into multiple video segments based on preset key content, and extracting key segments and non-key segments from the video segments, includes:

[0031] The key segment and the non-key segment are extracted from the video segment using a trained intelligent extraction model.

[0032] In some embodiments, uploading the key segment and the non-key segment according to the network status includes:

[0033] Based on network performance parameters, the network status is divided into multiple network levels;

[0034] Different numbers of upload threads are allocated to the key segments and the non-key segments according to the network level, so as to upload the key segments and the non-key segments.

[0035] In some embodiments, uploading the key segment and the non-key segment according to the network status includes:

[0036] Creating an upload progress record file for the key segment and the non-key segment;

[0037] If the uploading process is interrupted, the upload progress record file is read, and the key segment and the non-key segment are continuously uploaded from the interruption point according to the upload record file.

[0038] In some embodiments, uploading the key segment and the non-key segment according to the network status includes:

[0039] assigning different priorities to the key segments and the non-key segments according to their importance;

[0040] According to the different priorities, the key segments and the non-key segments are uploaded in order of priority.

[0041] In some embodiments, uploading the key segment and the non-key segment according to the network status further includes:

[0042] Performing integrity checks on the key segments and the non-key segments to check whether any of the key segments and the non-key segments are lost;

[0043] The key segments and the non-key segments are integrated to splice the scattered key segments and the non-key segments into the complete interview video.

[0044] In some embodiments, uploading the key segment and the non-key segment according to the network status further includes:

[0045] Conducting intelligent follow-up questions on the interviewee based on the key segments to generate follow-up video segments;

[0046] The follow-up video clip is uploaded according to the network status.

[0047] In some embodiments, the method of dividing the interview video into multiple video segments according to preset key content and extracting key segments and non-key segments from the video segments may also include:

[0048] Perform real-time detection of user device performance parameters and current network status, generate optimization suggestions and provide feedback to the user interface.

[0049] In a second aspect, the present disclosure provides an interview video uploading device, the interview video uploading device comprising:

[0050] A video processing module is configured to segment the interview video into a plurality of video segments according to preset key content, and extract key segments and non-key segments from the video segments; the key segments correspond to the preset key content;

[0051] The video uploading module is configured to upload the key segment and the non-key segment according to the network status.

[0052] In some embodiments, the interview video uploading device further includes:

[0053] a verification module configured to perform integrity verification on the key segments and the non-key segments to check whether any of the key segments and the non-key segments are lost;

[0054] The integration module is configured to integrate the key segments and the non-key segments to splice the scattered key segments and the non-key segments into the complete interview video.

[0055] In some embodiments, the interview video uploading device further includes:

[0056] an intelligent follow-up questioning module configured to conduct intelligent follow-up questions on the interviewee based on the key segments to generate follow-up questioning video segments;

[0057] The video uploading module is further configured to upload the questioning video clip according to the network status.

[0058] In some embodiments, the interview video uploading device further includes:

[0059] The real-time monitoring module is configured to perform real-time detection on the performance parameters of user equipment and the current network status, generate optimization suggestions and feed them back to the user interface.

[0060] In a third aspect, an embodiment of the present disclosure provides a computer device, the computer device comprising:

[0061] one or more processors;

[0062] a memory for storing one or more programs;

[0063] When the one or more programs are executed by the one or more processors, the one or more processors implement the interview video uploading method provided in the first aspect;

[0064] One or more I / O interfaces are connected between the processor and the memory and are configured to implement information interaction between the processor and the memory.

[0065] In a fourth aspect, an embodiment of the present disclosure provides a computer non-volatile readable storage medium, on which a computer program is stored. When the computer program is run by a processor, the interview video uploading method provided in the first aspect is executed. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 A flowchart of a method for uploading an interview video provided in an embodiment of the present disclosure.

[0067] Figure 2 A flowchart of the first interview video segmentation method provided in an embodiment of the present disclosure.

[0068] Figure 3 A flowchart of the second interview video segmentation method provided in an embodiment of the present disclosure.

[0069] Figure 4 A flowchart of the third interview video segmentation method provided in an embodiment of the present disclosure.

[0070] Figure 5 A flowchart of the fourth interview video segmentation method provided in an embodiment of the present disclosure.

[0071] Figure 6 A flowchart of the fifth interview video segmentation method provided in an embodiment of the present disclosure.

[0072] Figure 7 A flowchart of the sixth interview video segmentation method provided in an embodiment of the present disclosure.

[0073] Figure 8 A flowchart of the seventh interview video segmentation method provided in an embodiment of the present disclosure.

[0074] Figure 9 A flowchart of the first upload strategy provided for the implementation of the present disclosure.

[0075] Figure 10 A flowchart of the second upload strategy provided for the implementation of the present disclosure.

[0076] Figure 11 A flowchart of the third upload strategy provided for the implementation of the present disclosure.

[0077] Figure 12 A structural diagram of an interview video upload transposition method provided in an embodiment of the present disclosure.

[0078] Figure 13 This is a schematic diagram of the structure of a computer device provided in some embodiments of the present disclosure. DETAILED DESCRIPTION

[0079] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the present disclosure for protection, but merely represents selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present disclosure.

[0080] Unless otherwise defined, the technical or scientific terms used in this disclosure should have the usual meanings understood by people with ordinary skills in the field to which this disclosure belongs. The words "first", "second" and similar words used in this disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one", "an" or "the" do not indicate a quantity limitation, but rather indicate the existence of at least one. Words such as "include" or "comprise" mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0081] In the existing AI interview system, interviewees need to go through multiple interview question-and-answer sessions. The interview system will record multiple video clips based on the answers to multiple questions, and upload the recorded video clips to the interview management background in stages. It will also conduct real-time intelligent analysis of the interviewee's answer tendencies to questions to decide whether to initiate intelligent follow-up questioning. Finally, after the interview is completed, an evaluation report will be generated using the AI ​​big model.

[0082] However, due to many factors such as unstable network, poor performance of user equipment or outdated interview system version, there will be long waiting time, freezes or interruptions during the interview video upload process, resulting in unsuccessful submission of the interview paper. The interviewee is forced to re-interview and re-record the video, which also affects the accuracy of the professional interviewer's judgment in initiating intelligent follow-up questions, seriously affecting the user experience and interview efficiency.

[0083] In order to at least solve the above-mentioned technical problems, the embodiments of the present disclosure provide an interview video uploading method, device, computer equipment and storage medium. The interview video uploading method, device, computer equipment and storage medium provided by the embodiments of the present disclosure will be further described in detail below in combination with the accompanying drawings and specific implementation methods.

[0084] In a first aspect, the present disclosure provides a method for uploading an interview video. Figure 1 A flowchart of a method for uploading an interview video provided by an embodiment of the present disclosure is shown as follows: Figure 1 As shown, the interview video uploading method includes the following steps S101 to S102.

[0085] Step S101, according to preset key content, the interview video is divided into multiple video segments, and key segments and non-key segments are extracted from the video segments; the key segments correspond to the preset key content.

[0086] In step S101, the preset key content may be content closely related to the interview question, such as the interviewee's voice changes, facial expressions, movements, and important work experience and professional skills that directly reflect whether the interviewee is qualified for the current interview position. If multiple preset key contents appear in the interview video, the one with the higher weight is selected.

[0087] It should be noted here that professional interviewers can pre-mark different weights for preset key contents according to the field of the interview. For example, if the interview field is computer science, sentences containing technical terms such as "programming language", "algorithm design", and "project development process" have higher weights. If the above-mentioned technical terms appear in different preset key contents in the same interview video, the weights of the preset key contents corresponding to different technical terms can be sorted. For example, the preset key content corresponding to "programming language" has the highest weight, the preset key content corresponding to "algorithm design" has the second highest weight, and the preset key content corresponding to "project development process" has the lowest weight. Different weights can be set for the preset key contents corresponding to different technical terms in a certain field according to actual needs.

[0088] The interview video can be divided into multiple video segments according to key segments and non-key segments, where the key segments are video segments corresponding to preset key content, and the non-key segments are other video segments in the interview video except the key segments.

[0089] Step S102: uploading key segments and non-key segments according to the network status.

[0090] In step S102, the key and non-key segments are uploaded based on the network status. For example, when the network status is good, the key and non-key segments are uploaded simultaneously. When the network status is poor, the key segments are uploaded first. When the network status becomes good, the non-key segments are uploaded. Ultimately, all key and non-key segments in the interview video are uploaded. Alternatively, when the network status is poor, all key segments are uploaded first.

[0091] During the entire interview video upload process, key clips can directly impact the interview outcome. These clips are mandatory and can be uploaded first, without the interviewee's confirmation. Non-critical clips have a smaller impact on the interview outcome. Considering factors like the waiting time during the upload process, the interviewee can decide whether to upload the remaining non-critical clips after the key clips are uploaded, to avoid impacting the interview due to missed key clips.

[0092] In the interview video uploading method provided by the embodiment of the present disclosure, the interview video can be divided into multiple video segments according to key segments and non-key segments based on preset key content, so that key segments can be uploaded first according to network status during the uploading process. This can effectively solve the problems of video upload interruption, long user waiting time and unsuccessful interview submission caused by factors such as network changes or device performance during the AI ​​interview process, thereby improving the interview efficiency and user experience of the interviewer.

[0093] In some embodiments, Figure 2 This is a flow chart of the first interview video segmentation method provided by the embodiment of the present disclosure, as shown in FIG. Figure 2 As shown, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments, specifically including steps S201 to S203.

[0094] Step S201: using a speech recognition engine to recognize the speech content in the video clip and convert the speech content into text content.

[0095] In the above step S201, during the interview video recording process, the AI ​​interview system starts the voice recognition engine, recognizes the voice content in the recorded interview video in real time, and converts the voice content into text content.

[0096] In step S202 , natural language processing technology is used to perform semantic analysis on the text content, and natural language processing models for different positions are used to identify key texts in the text content that are relevant to the corresponding position.

[0097] In the above step S202, the text content is deeply semantically analyzed through natural language processing (NLP) technology, and pre-trained NLP models for different positions are used to identify keywords, key sentences and topics closely related to the job position. For example, for interview videos for technical positions, the NLP model will focus on sentences containing technical terms such as "programming language", "algorithm design" and "project development process". In actual applications, professional and technical terms can be input into the general model for training to form an NLP model for specific positions, so as to save model development and training costs.

[0098] Step S203: Determine the range of the corresponding video segment according to the timestamp corresponding to the key text on the timeline of the interview video, mark the video segment as the key segment, and mark the remaining video segments as non-key segments.

[0099] In step S203, the corresponding video segment range is determined based on the timestamp corresponding to the key text on the timeline of the interview video, marked as a key segment, and added to the priority upload queue. For example, if the text mentions "I was responsible for the design and implementation of the core algorithm in [XX project], and adopted [XX algorithm], which improved the project performance by [X%]", and the time corresponding to this text is from the 2nd minute to the 2nd minute and 30 seconds, then the video segment in this time period is the key segment, and the video segments in the same time period are non-key segments.

[0100] In some implementations, Figure 3 This is a flow chart of the second interview video segmentation method provided in the embodiment of the present disclosure, such as Figure 3 As shown, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments, specifically including steps S301 to S302.

[0101] Step S301: Using an image recognition algorithm, recognize the interviewee's facial expressions and body movements in each frame of image, and mark the key facial expressions and key body movements.

[0102] In step S301, an image recognition algorithm is used to analyze each frame of the interview video. The algorithm detects the interviewee's facial expressions, identifying states of confidence, nervousness, and confusion through facial landmark location and expression classification models. The algorithm also analyzes the interviewee's body language, such as posture and gestures, to determine whether they are emphasizing key points or expressing a specific attitude. For example, when answering important questions, the interviewee may lean forward, move their hands rhythmically, and display a confident expression.

[0103] Step S302, according to the timestamps corresponding to the images of the key facial expressions and key body movements on the timeline of the interview video, determine the range of the corresponding video segment, mark the video segment as the key segment, and mark the remaining video segments as non-key segments.

[0104] In step S302, the interview system records the timestamps of the specific facial expressions and body movements, and combines them with the time information of the voice content to determine that the video clip containing these key facial expressions, key body movements, and voice responses is a key clip. For example, if the interviewee says, "I am confident in this position and I have the ability to solve various complex problems," while leaning forward with a confident gesture, and the corresponding time is from the 8th minute to the 8th minute and 20 seconds, then the video clip within this time period is a key clip, and the video clips within the same time period are non-key clips.

[0105] In some implementations, Figure 4 This is a flow chart of the third interview video segmentation method provided by the embodiment of the present disclosure, such as Figure 4 As shown, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments, specifically including steps S401 to S402.

[0106] Step S401 , classifying pre-stored interview questions into key questions and non-key questions according to the types of interview questions.

[0107] In step S401 above, the AI ​​interview system pre-stores common interview question templates and corresponding answer extraction rules. For example, the question "Please briefly describe your work experience" is extremely important for determining whether the interviewee is qualified for the current interview position. This question can be classified as a key question, while other questions can be classified as non-key questions.

[0108] Step S402, according to the corresponding timestamps on the timeline of the interview video from the asking of the key question to the end of the answer, determine the range of the corresponding video segment, mark the video segment as the key segment, and mark the remaining video segments as non-key segments.

[0109] In step S402, the AI ​​interview system defines the video clip from the time the interviewee begins speaking after the question is asked to the time the interviewee finishes answering and there is a pause (e.g., 3 seconds) or a clear topic change sign (e.g., "Next, I want to talk about another experience of mine") as the key clip, corresponding to the 9th minute to 9 minutes and 40 seconds. The video clips in the remaining time periods are non-key clips. The AI ​​interview system automatically matches the corresponding template based on the order and type of questions in the interview process, extracting them as key clips and the remaining video clips as non-key clips.

[0110] In some implementations, Figure 5 This is a flow chart of the fourth interview video segmentation method provided in the embodiment of the present disclosure, such as Figure 5 As shown, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments, specifically including steps S501 to S502.

[0111] Step S501 : Classify pre-stored job skills into key job skills and non-key job skills according to the types of job skills.

[0112] In step S501, the AI ​​interview system creates a specialized template library for matching job skills and qualities to the requirements of different positions. For example, for a design position, the template will include key information extraction rules for design software proficiency, creative expression ability, and portfolio presentation.

[0113] Step S502: determine the range of the corresponding video segment based on the timestamp corresponding to the interviewee's description of key job skills on the timeline of the interview video, and mark the video segment as a key segment, and mark the remaining video segments as non-key segments.

[0114] In the above step S502, when the interviewee mentioned in the video "I am proficient in Adobe Photoshop and Illustrator, and recently completed the visual design of a [XX project], in which I used [XX design concept or technique]" and showed the relevant work screen, the corresponding time is from 10 minutes and 20 seconds to 10 minutes and 40 seconds. The AI ​​interview system recognized that this part of the video contained key job skill information based on the template, extracted it as a key segment, and the remaining video segments were non-key segments.

[0115] In some embodiments, Figure 6 This is a flow chart of the fifth interview video segmentation method provided by the embodiment of the present disclosure, such as Figure 6 As shown, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments, specifically including steps S601 to S602.

[0116] Step S601: using eye tracking technology to collect the interviewee's eye movement data and mark key eye movement data.

[0117] Step S602: Determine the range of the corresponding video segment based on the timestamp corresponding to the key eye movement data on the timeline of the interview video, mark the video segment as the key segment, and mark the remaining video segments as non-key segments.

[0118] In the above steps S601 to S602, if the AI ​​interview system is equipped with an eye tracking function, during the interview recording process, the AI ​​interview system collects the interviewee's eye movement data in real time, including gaze point coordinates, gaze time and other information. When the interviewee answers questions, the concentrated area and change trend of his or her gaze point are analyzed. For example, if the interviewee's eyes are fixed on the screen area showing the project results for a long time when he or she is talking about the project experience, the AI ​​interview system determines that the video clip containing this time period is a key clip based on the gaze time and the relevant timestamps, because this may imply that the interviewee is emphasizing the relationship between this part of the content and the project. For example, the video clip corresponds to the time from 12 minutes and 10 seconds to 12 minutes and 40 seconds, and the AI ​​interview system extracts it as a key clip, and the remaining video clips are non-key clips.

[0119] In some embodiments, Figure 7 This is a flow chart of the sixth interview video segmentation method provided in the embodiment of the present disclosure, such as Figure 7 As shown, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments, specifically including steps S701 to S702.

[0120] Step S701: Obtain the interviewee's human-computer interaction data and mark key human-computer interaction data.

[0121] Step S702: determine the range of the corresponding video segment according to the timestamp corresponding to the key human-computer interaction data on the timeline of the interview video, mark the video segment as the key segment, and mark the remaining video segments as non-key segments.

[0122] In the above steps S701 to S702, during the interview process, if there is an interactive link, such as the interviewee clicking a screen button to display supplementary materials (such as a scanned copy of a certificate, a project demonstration video, etc.), the AI ​​interview system records the time when the button is clicked and the time when the display is completed, and marks the video clips between these two time points as key clips, because these video clips contain supplementary information actively provided by the interviewee, which is of great value for the comprehensive evaluation of the interviewee. The remaining video clips can be marked as non-key clips.

[0123] In some embodiments, Figure 8 This is a flow chart of the seventh interview video segmentation method provided in the embodiment of the present disclosure, such as Figure 8 As shown, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments, specifically including steps S801 to S802.

[0124] Step S801: Obtain the interviewee's stress response data and mark key stress response data.

[0125] Step S802: determine the range of the corresponding video segment according to the timestamp corresponding to the key stress response data on the time axis of the interview video, mark the video segment as the key segment, and mark the remaining video segments as non-key segments.

[0126] In steps S801 to S802 above, if the AI ​​interview system is equipped with other sensors, it can detect the interviewee's stress response data, such as a pressure sensor (detecting changes in the interviewee's hand pressure to reflect tension) or a heart rate sensor (monitoring changes in heart rate to reflect emotional excitement), and integrate these sensor data with the interview video for analysis. For example, if the sensor data shows a sudden increase in the interviewee's emotional tension, these stress responses are key stress response data, and if the interviewee is answering a difficult question in the interview video, the video clip of this time period may be considered a key clip because it can reflect the interviewee's true performance under pressure. The remaining video clips can be marked as non-key clips.

[0127] In some embodiments, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments. Specifically, it may include: using a trained intelligent extraction model to extract key segments and non-key segments from the video segments.

[0128] A large amount of interview video data from different positions and types of interviewees is collected, and professional interviewers manually annotate key segments. This annotation is based on multiple factors, including the importance of the interviewee's answers, relevance to the position, and performance quality. The annotated data is divided into training, validation, and test sets. The training set data is used to train an intelligent extraction model, such as a support vector machine (SVM) or a deep neural network model. Input features for the intelligent extraction model include speech features (such as speech rate, intonation, and pauses), image features (such as facial expressions, body movements, and image changes), text features (such as keyword frequency and text semantic coherence), and temporal information (such as the duration of the answer and the timing of key content). After multiple rounds of training and validation, the intelligent extraction model parameters are adjusted to achieve a high accuracy rate (e.g., above 85%) on the test set. The trained intelligent extraction model is then used to automatically extract key segments from new interview videos, improving video processing efficiency and accuracy.

[0129] In some implementations, Figure 9 A flow chart of the first upload strategy provided for the implementation of the present disclosure is as follows: Figure 9 As shown, the above step S102, uploading the key segments and non-key segments according to the network status, specifically includes steps S901 to S902.

[0130] Step S901: Divide the network status into multiple network levels according to network performance parameters.

[0131] In the above step S901, the network status is quickly checked immediately before the video clip is uploaded. Based on network performance parameters such as bandwidth, delay, and packet loss rate, the network is divided into four levels: excellent (bandwidth ≥ 5Mbps, delay ≤ 50ms, packet loss rate ≤ 1%), good (2Mbps ≤ bandwidth < 5Mbps, 50ms < delay ≤ 100ms, 1% < packet loss rate ≤ 3%), medium (1Mbps ≤ bandwidth < 2Mbps, 100ms < delay ≤ 200ms, 3% < packet loss rate ≤ 5%), and poor (bandwidth < 1Mbps, delay > 200ms, packet loss rate > 5%).

[0132] Step S902 : allocating different numbers of upload threads to the key segments and the non-key segments according to the network level, so as to upload the key segments and the non-key segments.

[0133] In the above step S902, when the network is at an excellent level, the multi-threaded parallel upload mechanism is started, and more upload threads (such as 5 threads) are allocated to key clips, and each thread uploads at a higher rate (such as 1Mbps / thread); for non-key clips, fewer threads (such as 2 threads) are allocated and the rate is slightly lower (such as 0.5Mbps / thread). When the network is good, the number of key clip threads is reduced to 3, and the rate is reduced to 0.8Mbps / thread, and the number of non-key clip threads is 1, and the rate is 0.3Mbps / thread. When the network is medium, only one key clip upload thread is opened with a rate of 0.5Mbps, and non-key clips are suspended from uploading until the network improves. When the network is poor, all non-key clip uploads are suspended, and only key clips with a very small size (such as less than 1MB) and extremely important are uploaded, such as short video clips that contain personal core skills demonstrations or key project experiences. At the same time, the network status is checked every 10 seconds. Once the network is improved to a medium level or above, the upload of other video clips is gradually resumed.

[0134] In some implementations, Figure 10 A flow chart of the second upload strategy provided for the implementation of the present disclosure is as follows: Figure 10 As shown, the above step S102, uploading the key segments and non-key segments according to the network status, specifically includes steps S1001 to S1002.

[0135] Step S1001: Create upload progress record files for key segments and non-key segments.

[0136] If the uploading process is interrupted, step S1002 is executed to read the upload progress record file, and according to the upload record file, the key segments and non-key segments are continuously uploaded from the interruption point.

[0137] In the above steps S1001 to S1002, when the uploading of video segments (key segments and non-key segments) begins, an upload progress record file is created for each video segment to record information such as the number of bytes uploaded and the starting position of the upload.

[0138] If an interruption occurs during the upload process (such as a sudden network disconnection or device failure), the AI ​​interview system will first read the progress record file during the next upload, and continue uploading from the interruption point based on the recorded information, avoiding re-uploading of the successfully transmitted part. During the upload process, each time a video clip is uploaded, the system calculates the hash value of the data and compares it with the hash value of the corresponding clip pre-stored on the server. If the hash values ​​are consistent, the data is complete and accurate; if not, the video clip is re-uploaded to ensure the integrity and accuracy of the uploaded data.

[0139] In some embodiments, Figure 11 A flow chart of the third upload strategy provided for the implementation of the present disclosure is as follows: Figure 11 As shown, the above step S102, uploading the key segments and non-key segments according to the network status, specifically includes steps S1101 to S1102.

[0140] Step S1101 : assigning different priorities to key segments and non-key segments according to their importance.

[0141] In step S1101, a video clip upload priority queue is established, and different priorities are assigned based on the importance of the video clips. Key clips (such as those that demonstrate core job skills, important project experience, and high-quality answers to key questions) are given high priority, while non-key clips are given low priority.

[0142] Step S1102 : uploading key segments and non-key segments in order of priority according to different priorities.

[0143] In step S1102, when uploading, clips in the high-priority queue are prioritized. The execution order of the priority queues is dynamically adjusted based on network conditions. For example, when network conditions are good, each video clip is uploaded sequentially according to its priority. When network conditions deteriorate, the upload of high-priority video clips is prioritized, and the upload of low-priority video clips is suspended. The upload of low-priority video clips is resumed after the network improves, ensuring that the most critical video clips are uploaded first, given limited network resources.

[0144] In some implementations, such as Figure 1 As shown, the above step S102 uploads the key segments and non-key segments according to the network status, and then further includes: step S103 to step S104.

[0145] Step S103: performing integrity check on the key segments and non-key segments to check whether any key segments and non-key segments are lost.

[0146] In step S103, during the video clip upload process, the uploaded video clips are checked for integrity at regular intervals (e.g., 30 seconds) to check for missing or damaged video clips. This check is performed by comparing the locally recorded list of video clips to be uploaded with the server-side list of video clips already received. If any discrepancies are found, the missing or damaged video clips are promptly re-uploaded.

[0147] Step S104: integrating the key segments and non-key segments to splice the scattered key segments and non-key segments into a complete interview video.

[0148] In the above step S104, when all video clips have been uploaded, the AI ​​interview system integrates the video clips on the server side and splices the scattered video clips into a complete video file in chronological order. If it is found that some clips have not been uploaded successfully when the interview paper is submitted, the AI ​​interview system will pop up a prompt box in the user interface to inform the interviewee of the clips that have not been uploaded successfully and the possible reasons (such as network problems or equipment failures), and provide solution suggestions, such as "Your network connection is unstable, resulting in some video clips not being uploaded successfully. Please check the network and click Retry. The system will automatically resume the upload from the interruption point."

[0149] In some embodiments, as Figure 1 As shown, the above step S102 uploads the key segments and non-key segments according to the network status, and then further includes: steps S105 to S106.

[0150] Step S105 , conducting intelligent follow-up questions on the interviewee based on the key segments to generate follow-up video segments.

[0151] Step S106: Upload the questioning video clip according to the network status.

[0152] In the above steps S105 to S106, during the AI ​​interview process, when the video clips of the answer to a certain question are being recorded, the video clips are uploaded to the server one after another and recorded. After the interviewee completes the answer, if the complete interview video has been uploaded, the AI ​​digital human will conduct an overall comprehensive analysis based on the complete interview video and then conduct follow-up intelligent questions; if the key clips have been uploaded, and some other video clips are still being uploaded slowly, in order to ensure the timeliness experience, the AI ​​digital human will conduct a preliminary analysis based on the core content of the extracted key clips and then conduct follow-up intelligent questions.

[0153] When the AI ​​interview system identifies missing key information or details requiring further understanding, it automatically triggers an intelligent follow-up questioning mechanism. For example, if an interviewee mentions using a new technology to solve a project problem during a technical position interview but fails to elaborate on the technical details, the system will automatically generate follow-up questions based on a pre-set technical position-specific follow-up template, such as, "Please describe in detail the specific principles, operational steps, and integration of [XX new technology] you used." Similarly, for sales positions, if an interviewee fails to mention customer feedback when describing a successful sales case, the system can ask, "After this sale, what specific customer feedback did you receive about the product and service?" These follow-up questions are presented to the interviewee via the AI ​​digital human in video, voice, and text formats, requiring them to continue recording their responses within a specified timeframe. Newly recorded video clips are processed using the same optimization methods for interview video uploads described above, ensuring an efficient and smooth follow-up process, further exploring the interviewee's abilities and potential, and providing employers with a more comprehensive and accurate interview assessment.

[0154] After the entire interview is completed, all recorded video clips are integrated on the server. If non-critical clips are not uploaded, the front-end will provide feedback to the user to choose whether to re-record. If you choose not to record, the integrated videos will be uploaded and updated in the background one after another, and the uploaded video content will be analyzed and evaluated using the AI ​​big model, and a detailed evaluation report will be automatically generated, including multi-dimensional evaluation results such as the interviewee's performance and professional skill level, for professional interviewers to view and preview.

[0155] In some embodiments, as Figure 1 As shown, the above step S101 divides the interview video into multiple video segments according to preset key content, and extracts key segments and non-key segments from the video segments. It also includes: step S101A, which performs real-time detection on the performance parameters and current network status of the user device, generates optimization suggestions and feeds them back to the user interface.

[0156] Before the interviewee starts answering questions and recording the interview video, the user's device performance and network status are automatically detected. If there are problems such as low performance of the user's device, too low system version or poor network conditions, the interviewee will be prompted with possible problems and optimization suggestions will be provided. Optimization suggestions include but are not limited to prompting users to replace higher-performance devices, upgrade the AI ​​interview system version, or change network settings (such as switching from Wi-Fi to mobile data). Continuously monitor network status including bandwidth, latency, packet loss rate, etc. Network testing tools or self-developed network detection modules can be used to periodically (such as every second) send probe packets to a specific server and analyze the response. If the network bandwidth is low, automatically switch to a better network (such as switching from Wi-Fi to mobile data) or adjust the video quality to adapt to the network. For example, further reduce the resolution and bit rate, and display the current network status and its impact on recording on the user interface.

[0157] In a second aspect, the present disclosure provides a device for uploading an interview video. Figure 12 A schematic diagram of a structure of an interview video upload transposition provided by an embodiment of the present disclosure, such as Figure 12 As shown, the interview video uploading device includes: a video processing module 1201, a video uploading module 1202, a verification module 1203, an integration module 1204, an intelligent questioning module 1205, and a real-time monitoring module 1206.

[0158] The video processing module 1201 is configured to divide the interview video into multiple video segments according to preset key content, and extract key segments and non-key segments from the video segments; the key segments correspond to the preset key content; the video uploading module 1202 is configured to upload key segments and non-key segments according to the network status.

[0159] like Figure 12 As shown, the interview video uploading device further includes: a verification module 1203 and an integration module 1204. The verification module 1203 is configured to perform integrity verification on key segments and non-key segments to check whether any key segments and non-key segments are missing; the integration module 1204 is configured to integrate the key segments and non-key segments to splice the scattered key segments and non-key segments into a complete interview video.

[0160] like Figure 12 As shown, the interview video uploading device further includes: an intelligent questioning module 1205. The intelligent questioning module 1205 is configured to conduct intelligent questioning of the interviewee based on key segments to generate questioning video segments; the video uploading module 1202 is further configured to upload the questioning video segments based on the network status.

[0161] like Figure 12As shown, the interview video uploading device further includes a real-time monitoring module 1206. The real-time monitoring module 1206 is configured to perform real-time detection on the performance parameters of the user equipment and the current network status, generate optimization suggestions and feed them back to the user interface.

[0162] It should be noted that the various modules in the above-mentioned interview video uploading device can be implemented in whole or in part by software, hardware, or a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above-mentioned modules to implement the interview video uploading method provided in any of the above-mentioned embodiments. The specific implementation method is the same as the implementation method of the interview video uploading method provided in any of the above-mentioned embodiments, and will not be repeated here.

[0163] In a third aspect, an embodiment of the present disclosure provides a computer device, Figure 13 This is a schematic diagram of the structure of a computer device provided in some embodiments of the present disclosure, such as Figure 13 As shown, the computer device includes: one or more processors 1301; a memory 1302, used to store one or more programs, when the one or more programs are executed by one or more processors, the one or more processors implement the interview video uploading method provided by any of the above embodiments; one or more I / O interfaces 1303, connected between the processor and the memory, and configured to realize information interaction between the processor and the memory.

[0164] Among them, the processor 1301 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 1302 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read-write interface) 1303 is connected between the processor 1301 and the memory 1302, and can realize information interaction between the processor 1301 and the memory 1302, including but not limited to a data bus (Bus), etc.

[0165] In some embodiments, the processor 1301 , the memory 1302 , and the I / O interface 1303 are connected to each other via a bus, and further connected to other components of the computing device.

[0166] In a fourth aspect, this embodiment provides a computer non-volatile readable storage medium, on which a computer program is stored. When the program is executed by a processor, the interview video uploading method provided in any of the above embodiments is implemented.

[0167] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a machine-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above-mentioned functions defined in the system of the present disclosure are executed.

[0168] It should be noted that the computer non-transitory readable medium shown in the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any non-transitory computer-readable storage medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the non-transitory computer-readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination thereof.

[0169] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architectures, functions and operations of the devices, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the aforementioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in an order different from that marked in the accompanying drawings. For example, two connected boxes can actually represent execution in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0170] The circuits or sub-circuits described in the embodiments of the present disclosure may be implemented in software or hardware. The described circuits or sub-circuits may also be provided in a processor. For example, they may be described as: a processor comprising: a receiving circuit and a processing circuit, wherein the processing module comprises a writing sub-circuit and a reading sub-circuit. The names of these circuits or sub-circuits do not, in certain circumstances, constitute limitations on the circuits or sub-circuits themselves. For example, a receiving circuit may also be described as "receiving a signal."

[0171] It is understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present disclosure, and the present disclosure is not limited thereto. Those skilled in the art may make various modifications and improvements without departing from the spirit and substance of the present disclosure, and such modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A method for uploading an interview video, characterized in that: The interview video uploading method includes: According to the preset key content, the interview video is divided into multiple video segments, and key segments and non-key segments are extracted from the video segments; the key segments correspond to the preset key content; The key segment and the non-key segment are uploaded according to the network status.

2. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments includes: Using a speech recognition engine, recognizing speech content in the video clip and converting the speech content into text content; Using natural language processing technology to perform semantic analysis on the text content, and using natural language processing models for different positions to identify key text in the text content that is relevant to the corresponding position; According to the timestamp corresponding to the key text on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

3. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments includes: Use image recognition algorithms to identify the interviewee's facial expressions and body movements in each frame, and mark key facial expressions and key body movements; According to the timestamps corresponding to the images in which the key facial expressions and the key body movements appear on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

4. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments includes: According to the type of interview questions, pre-stored interview questions are divided into key questions and non-key questions; According to the timestamps corresponding to the time from when the key question is asked to when the answer is completed on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

5. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments includes: According to the type of job skills, the pre-stored job skills are divided into key job skills and non-key job skills; According to the timestamp corresponding to the interviewee's description of the key job skills on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

6. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments includes: Use eye tracking technology to collect interviewees’ eye movement data and mark key eye movement data; According to the timestamp corresponding to the key eye movement data on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

7. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments includes: Obtain the interviewee's human-computer interaction data and mark the key human-computer interaction data; According to the timestamp corresponding to the key human-computer interaction data on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

8. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments includes: Obtain interviewees' stress response data and mark key stress response data; According to the timestamp corresponding to the key stress response data on the timeline of the interview video, the range of the corresponding video segment is determined, and the video segment is marked as a key segment, and the remaining video segments are marked as non-key segments.

9. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments includes: The key segment and the non-key segment are extracted from the video segment using a trained intelligent extraction model.

10. The interview video uploading method according to claim 1, characterized in that: The uploading of the key segment and the non-key segment according to the network status includes: Based on network performance parameters, the network status is divided into multiple network levels; Different numbers of upload threads are allocated to the key segments and the non-key segments according to the network level, so as to upload the key segments and the non-key segments.

11. The interview video uploading method according to claim 1, characterized in that: The uploading of the key segment and the non-key segment according to the network status includes: Creating an upload progress record file for the key segment and the non-key segment; If the uploading process is interrupted, the upload progress record file is read, and the key segment and the non-key segment are continuously uploaded from the interruption point according to the upload record file.

12. The interview video uploading method according to claim 1, characterized in that: The uploading of the key segment and the non-key segment according to the network status includes: assigning different priorities to the key segments and the non-key segments according to their importance; According to the different priorities, the key segments and the non-key segments are uploaded in order of priority.

13. The interview video uploading method according to claim 1, characterized in that: The step of uploading the key segment and the non-key segment according to the network status further includes: Performing integrity checks on the key segments and the non-key segments to check whether any of the key segments and the non-key segments are lost; The key segments and the non-key segments are integrated to splice the scattered key segments and the non-key segments into the complete interview video.

14. The interview video uploading method according to claim 1, characterized in that: The step of uploading the key segment and the non-key segment according to the network status further includes: Conducting intelligent follow-up questions on the interviewee based on the key segments to generate follow-up video segments; The follow-up video clip is uploaded according to the network status.

15. The interview video uploading method according to claim 1, characterized in that: The method of dividing the interview video into multiple video segments according to the preset key content and extracting key segments and non-key segments from the video segments also includes: Perform real-time detection of user device performance parameters and current network status, generate optimization suggestions and provide feedback to the user interface.

16. An interview video uploading device, characterized in that: The interview video uploading device includes: A video processing module is configured to segment the interview video into a plurality of video segments according to preset key content, and extract key segments and non-key segments from the video segments; the key segments correspond to the preset key content; The video uploading module is configured to upload the key segment and the non-key segment according to the network status.

17. The interview video uploading device according to claim 16, characterized in that: The interview video uploading device also includes: a verification module configured to perform integrity verification on the key segments and the non-key segments to check whether any of the key segments and the non-key segments are lost; The integration module is configured to integrate the key segments and the non-key segments to splice the scattered key segments and the non-key segments into the complete interview video.

18. The interview video uploading device according to claim 16, characterized in that: The interview video uploading device also includes: an intelligent follow-up questioning module configured to conduct intelligent follow-up questions on the interviewee based on the key segments to generate follow-up questioning video segments; The video uploading module is further configured to upload the questioning video clip according to the network status.

19. The interview video uploading device according to claim 16, characterized in that: The interview video uploading device also includes: The real-time monitoring module is configured to perform real-time detection on the performance parameters of user equipment and the current network status, generate optimization suggestions and feed them back to the user interface.

20. A computer device, characterized in that: The computer device comprises: one or more processors; a memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the interview video uploading method according to any one of claims 1 to 15; One or more I / O interfaces are connected between the processor and the memory and are configured to implement information interaction between the processor and the memory.

21. A computer non-transitory readable storage medium, characterized in that: The computer non-transitory readable storage medium stores a computer program, and when the computer program is executed by a processor, the interview video uploading method according to any one of claims 1 to 15 is executed.

Images