Live advertising video content optimization method and system based on artificial intelligence

Abstract

The present invention relates to the field of audio data processing technology, and in particular to a method and system for optimizing live advertising video content based on artificial intelligence, as follows: collecting user gaze point information during the live broadcast sales of a single commodity; calculating the user's attention concentration at a certain moment according to the cumulative moving distance of the gaze point and the time when the gaze point falls on the anchor in the previous period of time; obtaining the audio signal of the corresponding anchor, calculating the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time; calculating the user's average attention to the current goods on sale according to the user's attention concentration and the correlation; when the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding goods pops up, thereby recommending goods. The present invention can quickly and effectively promote the completion of transactions, avoiding the problem of losing customers due to the possibility that users may give up due to complicated operations.

Description

Technical Field

[0001] The present invention relates to the technical field of audio data processing, and in particular to a method and system for optimizing live advertising video content based on artificial intelligence. Background Art

[0002] Nowadays, live broadcasting, as a form of real-time interactive media, is widely used in entertainment, education, business, social and other fields. Among them, in live broadcasting with goods, the optimization of live broadcast advertising content can help improve the attractiveness of live broadcasting, enhance user participation, improve conversion rate, and greatly facilitate users and improve user satisfaction. Existing technologies often use artificial intelligence technology to optimize live broadcast advertising content, and can adopt corresponding strategies and methods for different business needs and target user groups to achieve the best marketing effect.

[0003] Nowadays, artificial intelligence technology is often used to optimize live broadcast advertising content. However, during the live broadcast, it is often difficult to timely determine the user's attention to the products recommended in the current live broadcast, resulting in the inability to make product recommendations in a timely manner. Summary of the invention

[0004] In order to solve the problem that it is difficult to obtain the user's attention to the product in time during the live broadcast process in the prior art, resulting in the inability to make product recommendations in time, the purpose of the present invention is to provide a live broadcast advertising video content optimization method and system based on artificial intelligence, and the technical solution adopted is as follows:

[0005] A method for optimizing live advertising video content based on artificial intelligence, the method comprising the following steps:

[0006] Real-time collection of user gaze point information during the live broadcast of a single product sales process;

[0007] The user's concentration at a certain moment is calculated based on the cumulative movement distance of the gaze point in the previous period of time and the time the gaze point is on the anchor;

[0008] Obtain the audio signal of the corresponding anchor in the previous period of time at a certain moment, and calculate the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time;

[0009] The average attention degree of the user to the currently sold product is calculated based on the obtained user attention concentration degree and the correlation between the user attention concentration degree and the audio signal strength of the anchor;

[0010] When the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding product will pop up to make product recommendations.

[0011] Preferably, the gaze point information includes the gaze point position and the corresponding time;

[0012] According to the change of the gaze point position at two different times, the movement distance of the gaze point is calculated.

[0013] Furthermore, the calculation of the user's attention concentration at a certain moment according to the cumulative moving distance of the gaze point and the time when the gaze point falls on the anchor in a period of time before the certain moment includes:

[0014] Get the time when the gaze point falls on the anchor;

[0015] Calculate the proportion of time that the gaze point is on the anchor in the period before a certain moment;

[0016] According to the moving distance of the fixation point at two different times, the cumulative moving distance of the fixation point in the period before a certain moment is calculated;

[0017] The user's attention concentration at a certain moment is calculated by combining the proportion of time the gaze point is focused on the anchor and the cumulative movement distance of the gaze point in the period before a certain moment.

[0018] Preferably, obtaining an audio signal of a corresponding anchor in a period of time before a certain moment, and calculating the correlation between the user's attention concentration and the anchor's audio signal strength in the period of time before a certain moment, includes:

[0019] According to the obtained user attention concentration and the corresponding time, a curve of the user attention concentration changing over time is obtained;

[0020] Obtaining a time period in which the user's concentration is greater than or equal to a preset concentration threshold;

[0021] Calculate the prominence of the host's audio signal in a period before the time period;

[0022] Combining the time period with the prominence of the audio signal, the correlation between the user's attention concentration and the host's audio signal strength in the previous period of time is calculated.

[0023] Furthermore, the calculating of the prominence of the host's audio signal in a period before the time period includes:

[0024] Taking the starting point of the time period as the base point, taking an audio data segment of a certain time length forward;

[0025] Calculate the average audio signal strength corresponding to the audio data segment;

[0026] Divide each audio data segment into a number of frames according to a preset frame length, and obtain the audio signal strength corresponding to each frame;

[0027] By calculating the difference in audio signal strength between two adjacent frames, a differential sequence of audio signal strength corresponding to the time period is obtained;

[0028] The prominence corresponding to the audio data segment is calculated by combining the obtained average audio signal strength, the differential sequence of the audio signal strength corresponding to the time period, and the number of frames.

[0029] Preferably, the average attention of the user to the currently sold product is calculated based on the obtained user attention concentration and the correlation between the user attention concentration and the anchor's audio signal strength, including:

[0030] Correcting the user's attention concentration according to the obtained user's attention concentration and the correlation between the user's attention concentration and the host's audio signal strength;

[0031] Get the time to start selling the product and the current time;

[0032] Calculate the modified sum of the user's attention concentration from the time when the product starts to be sold to the current time;

[0033] Calculate the length of time from the start of selling the product to the current time;

[0034] According to the modified sum of the user's attention concentration from the start time of selling the goods to the current time, and the length of time from the start time of selling the goods to the current time, the average attention of the user to the currently sold goods is calculated.

[0035] Preferably, while collecting the user's gaze point information in real time during the live broadcast sales of a single commodity, the method includes:

[0036] Collect the audio signal of the host at the corresponding time in real time.

[0037] A live broadcast advertising video content optimization system based on artificial intelligence, the system comprising:

[0038] The first acquisition module is used to collect the user's gaze point information in real time during the live broadcast sales of a single product;

[0039] The attention concentration calculation module is used to calculate the user's attention concentration at a certain moment based on the cumulative movement distance of the gaze point in the previous period of time and the time when the gaze point falls on the anchor;

[0040] The second acquisition module is used to obtain the audio signal of the corresponding anchor in the previous period of time at a certain moment;

[0041] A correlation calculation module, used to calculate the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time based on the corresponding anchor's audio signal in the previous period of time at a certain moment;

[0042] The average attention calculation module is used to calculate the average attention of the user to the currently sold product based on the obtained user attention concentration and the correlation between the user attention concentration and the audio signal strength of the anchor;

[0043] The product recommendation module, when the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding product will pop up, thereby making product recommendations.

[0044] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for optimizing live advertising video content based on artificial intelligence is implemented.

[0045] A computer-readable storage medium stores a computer program, which, when executed by a processor, enables the processor to execute the artificial intelligence-based live advertising video content optimization method.

[0046] The present invention has the following beneficial effects:

[0047] The present invention first calculates the user's attention concentration during the live broadcast by analyzing the cumulative moving distance of the gaze point and the time when the gaze point falls on the anchor, and then calculates the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time, thereby eliminating the influence of the anchor's audio signal strength on the user's attention concentration, and finally calculates the user's average attention to the currently sold goods. When the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding goods pops up to recommend goods. The present invention timely obtains the user's attention to the goods, thereby popping up a one-click purchase window. Through the method of the present invention, the advertising content is optimized, and when the user intends to buy, the transaction can be quickly and effectively promoted to complete, avoiding the problem of losing customers due to the user's possible abandonment due to complicated operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0049] Figure 1A flowchart of a method for optimizing live advertising video content based on artificial intelligence provided by an embodiment of the present invention.

[0050] Figure 2 A schematic diagram of a live advertising video content optimization system based on artificial intelligence provided by an embodiment of the present invention.

[0051] Figure 3 A functional block diagram of a computer device provided in accordance with an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a live advertising video content optimization method based on artificial intelligence proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0053] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0054] The following is a detailed description of a specific solution of a live advertising video content optimization method based on artificial intelligence provided by the present invention in conjunction with the accompanying drawings.

[0055] In real life, especially during live broadcast, it is often difficult to timely obtain the user's attention to the recommended products in the current live broadcast, which leads to the inability to make product recommendations in time, and the purchase operation is relatively complicated, resulting in the loss of users. Therefore, the present invention proposes a live broadcast advertising video content optimization method based on artificial intelligence, which mainly calculates the user's attention concentration during the live broadcast process, and eliminates the influence of the host's audio signal strength on the user's attention concentration, timely obtains the user's purchase intention, pops up a one-key purchase window, and improves the order transaction rate.

[0056] Please refer to Figure 1 , which shows a flow chart of a live advertisement video content optimization method based on artificial intelligence provided by an embodiment of the present invention, the method comprising the following steps:

[0057] Real-time collection of user gaze point information during the live broadcast of a single product sales process;

[0058] The user's concentration at a certain moment is calculated based on the cumulative movement distance of the gaze point in the previous period of time and the time the gaze point is on the anchor;

[0059] Obtain the audio signal of the corresponding anchor in the previous period of time at a certain moment, and calculate the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time;

[0060] The average attention degree of the user to the currently sold product is calculated based on the obtained user attention concentration degree and the correlation between the user attention concentration degree and the audio signal strength of the anchor;

[0061] When the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding product will pop up to make product recommendations.

[0062] The present invention first calculates the user's attention concentration during the live broadcast by analyzing the cumulative moving distance of the gaze point and the time when the gaze point falls on the anchor, and then calculates the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time, thereby eliminating the influence of the anchor's audio signal strength on the user's attention concentration, and finally calculates the user's average attention to the currently sold goods. When the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding goods pops up to recommend goods. The present invention timely obtains the user's attention to the goods, thereby popping up a one-click purchase window. Through the method of the present invention, the advertising content is optimized, and when the user intends to buy, the transaction can be quickly and effectively promoted to complete, avoiding the problem of losing customers due to the user's possible abandonment due to complicated operations.

[0063] In a specific embodiment, the user's gaze point information is collected in real time during the live broadcast of a single commodity, including: after the user agrees to collect the face during the live broadcast, the user's gaze point changes during the live broadcast are collected in real time through the front camera of the mobile phone. Specifically, the front camera can be used to obtain the user's face image, the face image is grayed and then input into FaceNet, and the real-time first image set and second image set are output. The first image set is to segment the facial triangle area of ​​each image, obtain the distance from the center of the eyeball to the center of the face, the distance from the center of the eyeball to the inner corner of the eye, and the angle from the center of the eyeball to the center of the face and the inner corner of the eye, and the eyeball includes the left eyeball and the right eyeball; the second image set is the numerical value of the facial area, head angle, and eye corner position of each image; the first image set is used as the input of the convolution layer of the pre-trained gaze point estimation model, and the second image set and the real-time user distance value and gaze boundary value obtained by the depth sensor are used as the input of the fully connected layer of the pre-trained gaze point estimation model, and the real-time gaze point estimation coordinates are output; the corresponding time is obtained at the same time, and the movement distance of the gaze point is calculated according to the change of the gaze point position at two different times. Other existing artificial intelligence technologies can also be used to obtain the changes in the user's gaze point during the live broadcast, as long as the gaze point position and corresponding time and other gaze point information can be collected. Since collecting the user's gaze point information during the live broadcast sales of a single product is not the invention point of the present invention, it will not be explained in detail here.

[0064] In this embodiment, the gaze point information includes the gaze point position and the corresponding time;

[0065] According to the change of the gaze point position at two different times, the movement distance of the gaze point is calculated.

[0066] In a specific embodiment, in order to calculate the user's attention to the products being sold during the current live broadcast, the existing artificial intelligence technology can be used to analyze the changes in the user's eye gaze point. First, the user's attention concentration on the mobile phone screen is calculated. The more concentrated the user's attention is on the mobile phone screen, the more likely the user is to pay attention to the products currently being sold. If a window for recommending products can be popped up in real time at this time, the user can purchase them with one click, which not only facilitates user operation but also achieves the best marketing effect.

[0067] The method of calculating the user's attention concentration at a certain moment according to the cumulative moving distance of the gaze point in a period of time before the certain moment and the time when the gaze point falls on the anchor includes:

[0068] Get the time when the gaze point falls on the anchor;

[0069] Calculate the proportion of time that the gaze point is on the anchor in the period before a certain moment;

[0070] According to the moving distance of the fixation point at two different times, the cumulative moving distance of the fixation point in the period before a certain moment is calculated;

[0071] The user's attention concentration at a certain moment is calculated by combining the proportion of time the gaze point is focused on the anchor and the cumulative movement distance of the gaze point in the period before a certain moment.

[0072] In this embodiment, the user's concentration at a certain moment is calculated by the cumulative movement distance of the gaze point and the time the gaze point is on the anchor in the period before the certain moment. For the cumulative movement distance of the gaze point, the larger the cumulative movement distance, the less focused the user is at this time, and the user is not very interested in the live broadcast product. For the time the user's gaze point is on the anchor, the longer the time, the more the user pays attention to the mobile phone live broadcast content, and the higher the user's concentration.

[0073] For a certain moment Calculate the user's attention concentration and analyze the current moment Some time ago The change of the gaze point within . The selection time cannot be too long. If the selected time period is too long, The user's attention concentration varies greatly within a period of time and has no reference value; of course, it cannot be too short, otherwise the data will not be representative. First, artificial intelligence technology is used to analyze the current user's gaze point during this period of time, and the time before the current moment is recorded. The cumulative moving distance of the user's gaze point is . Remember the time before the current moment The time that the user's gaze falls on the anchor is Based on this, the current time is calculated User attention span , the calculation formula is:

[0074]

[0075] In the formula, Indicates the current time Some time ago The proportion of time that the user's gaze is focused on the anchor. The larger the ratio, the more the user The greater the concentration at any moment. Indicates the current time Some time ago The cumulative moving distance of the user's gaze point. The smaller the cumulative moving distance, the less the user's eye movement, and the greater the user's concentration; Represents the normalization function.

[0076] In a specific embodiment, although the user's attention concentration at a certain moment is calculated, during the live broadcast, the host suddenly raises his voice, which may attract the user's attention, resulting in an increase in the user's attention concentration on the live broadcast process, making the user's attention to the product inaccurate. At this time, the user's attention concentration is closely related to the strength of the host's audio signal. Therefore, it is necessary to calculate the correlation between the user's attention concentration and the strength of the host's audio signal over a period of time; the stronger the correlation, the greater the possibility that the calculated user's attention concentration is disturbed by the host's voice, and the less it can reflect the user's attention to the current product. The weaker the correlation, the less likely the calculated user's attention concentration is to be disturbed by the host's voice, and the more it can reflect the user's attention to the current product.

[0077] The method of acquiring the audio signal of the corresponding host in a period of time before a certain moment and calculating the correlation between the user's attention concentration and the host's audio signal strength in the period of time before a certain moment in this embodiment includes:

[0078] According to the obtained user attention concentration and the corresponding time, a curve of the user attention concentration changing over time is obtained;

[0079] Obtaining a time period in which the user's concentration is greater than or equal to a preset concentration threshold;

[0080] Calculate the prominence of the host's audio signal in a period before the time period;

[0081] Combining the time period with the prominence of the audio signal, the correlation between the user's attention concentration and the host's audio signal strength in the previous period of time is calculated.

[0082] In this embodiment, by drawing a curve showing the change of user attention concentration over time, a preset attention concentration threshold is set (the threshold is set to 0.7 in this embodiment, and is set according to actual needs). Each curve segment exceeding the preset attention concentration threshold indicates a relatively high user attention concentration, and is suspected to have a high degree of attention to the product. Each segment of the change curve is used as a research object, and the intensity of the host's audio signal is compared on the same time axis, and the correlation between each segment of the user attention concentration change curve and the host's audio signal intensity is calculated.

[0083] The definition of correlation is: for each section of the attention concentration change curve, determine the prominence of the audio signal in the previous period of the section of the change curve; if the prominence is greater, it indicates that the increase in user attention concentration may be caused by the host's audio signal strength being too prominent, then the correlation is greater. And in this section of the change curve, the user's attention concentration drops significantly in a short period of time, indicating that the change in user attention concentration may be affected by the host's audio rather than the attention to the product itself, that is, the greater the correlation. Among them, the prominence of the audio signal is defined as: the greater the average audio signal strength, the more prominent the signal data; the greater the degree of change in the audio signal strength, the more prominent the signal data.

[0084] In a specific embodiment, while collecting the user's gaze point information during the live broadcast sales of a single product in real time, the method includes: collecting the host's audio signal at the corresponding time in real time. The real-time host's audio signal and the corresponding time can be stored in the cloud. When a user's attention concentration change curve exceeding a preset attention concentration threshold is obtained, the corresponding host's audio signal is extracted from the cloud with the same time axis.

[0085] In this embodiment, the calculating the prominence of the host's audio signal in a period before the time period includes:

[0086] Taking the starting point of the time period as the base point, taking an audio data segment of a certain time length forward;

[0087] Calculate the average audio signal strength corresponding to the audio data segment;

[0088] Divide each audio data segment into a number of frames according to a preset frame length, and obtain the audio signal strength corresponding to each frame;

[0089] By calculating the difference in audio signal strength between two adjacent frames, a differential sequence of audio signal strength corresponding to the time period is obtained;

[0090] The prominence corresponding to the audio data segment is calculated by combining the obtained average audio signal strength, the differential sequence of the audio signal strength corresponding to the time period, and the number of frames.

[0091] For each time period T when the user's attention is focused, the present invention believes that if the attention is attracted by audio factors, the host's audio signal will have certain abnormalities in a period of time before the time period T when the attention is focused, which is specifically manifested in that the average audio signal intensity in the audio data segment is strong and the overall intensity of the audio signal varies greatly. This time is set to TS=1s.

[0092] Therefore, for each period of focused attention , by time period The starting point is taken as the base point, and the length of the forward extraction in the audio signal is The audio data segment . Calculate the audio data segment Average audio signal strength within In addition, for the audio data segment Frame division is performed, and the frame length is set to 50ms. The audio signal strength corresponding to each frame is , calculate the audio data segment The difference in audio signal strength between two adjacent frames in are , and the average value of these values ​​represents the degree of change in the audio signal strength. The prominence is:

[0093]

[0094] In the formula, Indicates audio data segment The average audio signal strength within the audio data segment is greater than the average audio signal strength within the audio data segment. The greater the prominence. Indicates this audio data segment The mean value of the difference in audio signal strength between all two adjacent frames in the audio signal data segment. The larger the mean value, the stronger the audio signal data segment. The greater the change in The greater the prominence.

[0095] In this embodiment, the attention concentration time period is obtained Some time ago Corresponding audio data segment The prominence of the user is combined with whether the user's concentration has dropped significantly in a short period of time to calculate the concentration time period. The correlation with the intensity of the host's audio signal. Based on this, the time period of concentrated attention Correlation with the host's audio signal strength for:

[0096]

[0097] In the formula, Indicates the time period of concentration Some time ago Corresponding audio data segment The greater the prominence, the greater the possibility that the user's attention concentration is disturbed by the host's audio signal, and the greater the correlation between the user's attention concentration and the intensity of the host's audio signal. Indicates the time period of concentration. The shorter the time period of concentration, the more likely it is that the concentration is disturbed by the host's audio signal. The longer the user focuses on the current product due to non-subjective factors, the greater the correlation between the user's concentration and the intensity of the host's audio signal. Indicates the time period of concentration Correlation with the host's audio signal strength.

[0098] In a specific embodiment, the average attention of the user to the currently sold product is calculated based on the obtained user attention concentration and the correlation between the user attention concentration and the anchor's audio signal strength, including:

[0099] Correcting the user's attention concentration according to the obtained user's attention concentration and the correlation between the user's attention concentration and the host's audio signal strength;

[0100] Get the time to start selling the product and the current time;

[0101] Calculate the modified sum of the user's attention concentration from the time when the product starts to be sold to the current time;

[0102] Calculate the length of time from the start of selling the product to the current time;

[0103] According to the modified sum of the user's attention concentration from the start time of selling the goods to the current time, and the length of time from the start time of selling the goods to the current time, the average attention of the user to the currently sold goods is calculated.

[0104] In this embodiment, the average attention degree of the product is that the higher the user's attention concentration is, the higher the user's attention to the currently recommended product is. The higher the correlation between the user's attention concentration and the strength of the anchor's audio signal is, the lower the user's attention to the currently recommended product is. The time is , the current time is Based on this, users have a better understanding of the products currently on sale. Average attention for:

[0105]

[0106] In the formula, express The user’s attention concentration during the live broadcast; Indicates the corrected concentration after eliminating the interference of the host's audio signal strength on the user's concentration; Indicates the products currently on sale The average attention during the entire sales process, which indicates the user's current purchasing intention. Indicates that the product is starting to be sold Time To the current time For the time period Integrate and sum, that is, the corrected concentration is arrive The sum over the time period.

[0107] In this embodiment, after obtaining the average attention of the product currently sold by the user, Afterwards, it is determined whether the average attention is greater than or equal to a preset attention threshold (set as ). When a user Average attention When the user purchases The desire for the product is relatively strong, and a one-click purchase pops up at this time The window of goods not only facilitates users but also prevents merchants from missing out on users.

[0108] Based on the above-mentioned method for optimizing live advertisement video content based on artificial intelligence, this embodiment also provides a system for optimizing live advertisement video content based on artificial intelligence, such as Figure 2 As shown, the system comprises:

[0109] The first acquisition module is used to collect the user's gaze point information in real time during the live broadcast sales of a single product;

[0110] The attention concentration calculation module is used to calculate the user's attention concentration at a certain moment based on the cumulative movement distance of the gaze point in the previous period of time and the time when the gaze point falls on the anchor;

[0111] The second acquisition module is used to obtain the audio signal of the corresponding anchor in the previous period of time at a certain moment;

[0112] A correlation calculation module, used to calculate the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time based on the corresponding anchor's audio signal in the previous period of time at a certain moment;

[0113] The average attention calculation module is used to calculate the average attention of the user to the currently sold product based on the obtained user attention concentration and the correlation between the user attention concentration and the audio signal strength of the anchor;

[0114] The product recommendation module, when the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding product will pop up, thereby making product recommendations.

[0115] See also Figure 3 , Figure 3 5 is a schematic block diagram of a computer device provided by an embodiment of the present invention. The computer device 500 is a server, which may be an independent server or a server cluster composed of multiple servers.

[0116] See also Figure 3 The computer device 500 includes a processor 502 , a memory and a network interface 505 connected via a system bus 501 , wherein the memory may include a non-volatile storage medium 503 and an internal memory 504 .

[0117] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. When the computer program 5032 is executed, the processor 502 can execute a live advertisement video content optimization method based on artificial intelligence.

[0118] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500 .

[0119] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can execute the live advertising video content optimization method based on artificial intelligence.

[0120] The network interface 505 is used for network communication, such as providing data information transmission, etc. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device 500 to which the solution of the present invention is applied. The specific computer device 500 may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0121] The processor 502 is used to run a computer program 5032 stored in the memory to implement the artificial intelligence-based live advertising video content optimization method disclosed in an embodiment of the present invention.

[0122] Those skilled in the art will understand that Figure 3 The embodiments of the computer device shown in the figure do not constitute a limitation on the specific composition of the computer device. In other embodiments, the computer device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. For example, in some embodiments, the computer device may only include a memory and a processor. In such embodiments, the structure and function of the memory and the processor are the same as those of the embodiment of the present invention. Figure 3 The embodiments shown are consistent and will not be described again here.

[0123] It should be understood that in the embodiment of the present invention, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0124] In another embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium may be a non-volatile computer-readable storage medium. The computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, the method for optimizing live advertising video content based on artificial intelligence disclosed in an embodiment of the present invention is implemented.

[0125] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the above-described equipment, devices and units can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. Those of ordinary skill in the art can appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0126] In the several embodiments provided by the present invention, it should be understood that the disclosed equipment, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. Units with the same function may also be combined into one unit. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices or units, or may be an electrical, mechanical or other form of connection.

[0127] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present invention.

[0128] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0129] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), disk or optical disk and other media that can store program codes.

[0130] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0131] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for optimizing live advertising video content based on artificial intelligence, characterized in that: The method comprises the following steps: Real-time collection of user gaze point information during the live broadcast of a single product sales process; The user's concentration at a certain moment is calculated based on the cumulative movement distance of the gaze point in the previous period of time and the time the gaze point is on the anchor; Obtain the audio signal of the corresponding anchor in the previous period of time at a certain moment, and calculate the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time; The average attention degree of the user to the currently sold product is calculated based on the obtained user attention concentration degree and the correlation between the user attention concentration degree and the audio signal strength of the anchor; When the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding product will pop up, thereby making product recommendations; Among them, users have a certain understanding of the products currently on sale. Average attention for: In the formula, express The user’s attention concentration during the live broadcast; Indicates the corrected concentration after eliminating the interference of the host's audio signal strength on the user's concentration; Indicates the products currently on sale The average attention during the entire sales process, which indicates the user's current purchase intention. Indicates that the product is starting to be sold Time To the current time For the time period Integrate and sum, that is, the corrected concentration is arrive The sum over the time period.

2. The method for optimizing live advertising video content based on artificial intelligence according to claim 1, characterized in that: The gaze point information includes the gaze point position and the corresponding time; According to the change of the gaze point position at two different times, the movement distance of the gaze point is calculated.

3. The method for optimizing live advertising video content based on artificial intelligence according to claim 2, characterized in that: The method of calculating the user's attention concentration at a certain moment according to the cumulative moving distance of the gaze point in a period of time before the certain moment and the time when the gaze point falls on the anchor includes: Get the time when the gaze point falls on the anchor; Calculate the proportion of time that the gaze point is on the anchor in the period before a certain moment; According to the moving distance of the fixation point at two different times, the cumulative moving distance of the fixation point in the period before a certain moment is calculated; The user's attention concentration at a certain moment is calculated by combining the proportion of time the gaze point is focused on the anchor and the cumulative movement distance of the gaze point in the period before a certain moment.

4. The method for optimizing live advertising video content based on artificial intelligence according to claim 1, characterized in that: Obtain the audio signal of the corresponding anchor in the previous period of time at a certain moment, and calculate the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time, including: According to the obtained user attention concentration and the corresponding time, a curve of the user attention concentration changing over time is obtained; Obtaining a time period in which the user's concentration is greater than or equal to a preset concentration threshold; Calculate the prominence of the host's audio signal in a period before the time period; Combining the time period with the prominence of the audio signal, the correlation between the user's attention concentration and the host's audio signal strength in the previous period of time is calculated.

5. The method for optimizing live advertisement video content based on artificial intelligence according to claim 4, characterized in that: The calculating the prominence of the host's audio signal in a period before the time period includes: Taking the starting point of the time period as the base point, taking an audio data segment of a certain time length forward; Calculate the average audio signal strength corresponding to the audio data segment; Divide each audio data segment into a number of frames according to a preset frame length, and obtain the audio signal strength corresponding to each frame; By calculating the difference in audio signal strength between two adjacent frames, a differential sequence of audio signal strength corresponding to the time period is obtained; The prominence corresponding to the audio data segment is calculated by combining the obtained average audio signal strength, the differential sequence of the audio signal strength corresponding to the time period, and the number of frames.

6. The method for optimizing live advertising video content based on artificial intelligence according to claim 1, characterized in that: While collecting the user's gaze point information in real time during the live broadcast sales of a single commodity, the method includes: Collect the audio signal of the host at the corresponding time in real time.

7. A live advertising video content optimization system based on artificial intelligence, characterized by: The system comprises: The first acquisition module is used to collect the user's gaze point information in real time during the live broadcast sales of a single product; The attention concentration calculation module is used to calculate the user's attention concentration at a certain moment based on the cumulative movement distance of the gaze point in the previous period of time and the time when the gaze point falls on the anchor; The second acquisition module is used to obtain the audio signal of the corresponding anchor in the previous period of time at a certain moment; A correlation calculation module, used to calculate the correlation between the user's attention concentration and the anchor's audio signal strength in the previous period of time based on the corresponding anchor's audio signal in the previous period of time at a certain moment; The average attention calculation module is used to calculate the average attention of users to the currently sold products based on the obtained user attention concentration and the correlation between the user attention concentration and the anchor's audio signal strength. Average attention for: In the formula, express The user’s attention concentration during the live broadcast; Indicates the corrected concentration after eliminating the interference of the host's audio signal strength on the user's concentration; Indicates the products currently on sale The average attention during the entire sales process, which indicates the user's current purchase intention. Indicates that the product is starting to be sold Time To the current time For the time period Integrate and sum, that is, the corrected concentration is arrive The sum of the time periods; The product recommendation module, when the average attention is greater than or equal to the preset attention threshold, a window for purchasing the corresponding product will pop up, thereby making product recommendations.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the live advertising video content optimization method based on artificial intelligence as described in any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, enables the processor to execute the artificial intelligence-based live advertising video content optimization method as described in any one of claims 1 to 6.

Images