User evaluation method, device, equipment and computer-readable storage medium

Abstract

The present invention provides a user evaluation method, device, equipment and computer-readable storage medium, the user evaluation method comprising: obtaining audio and video information of a user during a target service process; extracting features of audio and video clips corresponding to each service link in the audio and video information in turn according to the service links included in the target service, and obtaining feature information of each service link; analyzing the feature information of each service link in turn, and obtaining the user's evaluation result of each service link. The embodiment of the present invention can obtain the user's partial evaluation of a single link of the service in a precise manner, thereby fully understanding the user's evaluation of the service content, improving the accuracy of the evaluation result, and helping to optimize and improve the service quality.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a user evaluation method, device, equipment and computer-readable storage medium. Background Art

[0002] At present, when implementing user evaluation, users are usually asked to give evaluations through text messages, electronic questionnaires, etc. after receiving the service. For example, after providing business services, the communication business hall usually sends a text message to the user to ask the user to provide an evaluation of the service. However, this user evaluation method usually only requires users to give a comprehensive evaluation of the service, which is not detailed enough and cannot fully understand the user's evaluation of the service content. Summary of the invention

[0003] The embodiments of the present invention provide a user evaluation method, device, equipment and computer-readable storage medium to solve the problem that the existing user evaluation method cannot fully understand the user's evaluation of service content.

[0004] In order to solve the above-mentioned technical problems, the present invention is achieved as follows:

[0005] In a first aspect, an embodiment of the present invention provides a user evaluation method, which is applied to a user evaluation device, comprising:

[0006] Obtain the user's audio and video information during the target service process;

[0007] According to the service links included in the target service, feature extraction is performed on the audio and video clips corresponding to each service link in the audio and video information in turn to obtain feature information of each service link;

[0008] The characteristic information of each service link is analyzed in turn to obtain the user's evaluation result of each service link.

[0009] Optionally, analyzing the characteristic information of each service link in turn to obtain the user's evaluation result of each service link includes:

[0010] The characteristic information of each service link is input into the pre-trained evaluation model of each service link in turn to obtain the user's evaluation result of each service link.

[0011] Optionally, before obtaining the audio and video information of the user during the target service, the method further includes:

[0012] Get the service training collection;

[0013] Classifying the services in the service training set to obtain each category of services;

[0014] According to a preset rule, each type of service is divided into N service links, where N is an integer greater than or equal to 1;

[0015] For each service link of each type of service, the following processes are performed respectively:

[0016] Respectively obtaining a plurality of audio and video clips of a plurality of users corresponding to the service link, and a plurality of evaluation results of the plurality of users on the service link;

[0017] Extracting features from the multiple audio and video clips to obtain multiple feature information of the service link;

[0018] An evaluation model for the service link is trained by utilizing the multiple feature information of the service link and the multiple evaluation results.

[0019] Optionally, according to the service links included in the target service, feature extraction is performed on the audio and video clips corresponding to each service link in the audio and video information in turn to obtain feature information of each service link, including:

[0020] Analyze the audio and video information, and determine each service link of the target service in sequence according to the analysis result and the service links included in the target service;

[0021] The audio and video clips corresponding to each of the service links are sequentially intercepted from the audio and video information, and features are extracted from the audio and video clips corresponding to each of the service links to obtain feature information of each of the service links.

[0022] Optionally, after obtaining the user's evaluation result for each of the service links, the method further includes:

[0023] According to the evaluation results of the user on each of the service links, the comprehensive evaluation results of the user on the target service are determined.

[0024] Optionally, after obtaining the user's evaluation result for each of the service links, the method further includes:

[0025] The audio and video clips corresponding to each service link and the user's evaluation results for each service link are stored in the database to serve as training samples to update the evaluation model of each service link.

[0026] Optionally, the feature information includes at least one of the following:

[0027] Facial expression feature information, body movement feature information and speech semantic feature information.

[0028] In a second aspect, an embodiment of the present invention provides a user evaluation apparatus, which is applied to a user evaluation device, including:

[0029] The first acquisition module is used to acquire the audio and video information of the user during the target service process;

[0030] A feature extraction module, configured to extract features from the audio and video clips corresponding to each service link in the audio and video information in turn according to the service links included in the target service, so as to obtain feature information of each service link;

[0031] The analysis module is used to analyze the characteristic information of each service link in turn to obtain the user's evaluation result of each service link.

[0032] Optionally, the analysis module is specifically used for:

[0033] The characteristic information of each service link is input into the pre-trained evaluation model of each service link in turn to obtain the user's evaluation result of each service link.

[0034] Optionally, the user evaluation device may further include:

[0035] A second acquisition module is used to acquire a service training set;

[0036] A classification processing module, used for classifying the services in the service training set to obtain each type of service;

[0037] A division module, used to divide each type of service into N service links according to a preset rule, where N is an integer greater than or equal to 1;

[0038] The execution module is used to execute the following processes for each service link of each type of service:

[0039] Respectively obtaining a plurality of audio and video clips of a plurality of users corresponding to the service link, and a plurality of evaluation results of the plurality of users on the service link;

[0040] Extracting features from the multiple audio and video clips to obtain multiple feature information of the service link;

[0041] An evaluation model for the service link is trained by utilizing the multiple feature information of the service link and the multiple evaluation results.

[0042] Optionally, the feature extraction module may include:

[0043] An analysis unit, configured to analyze the audio and video information, and determine each service link of the target service in sequence according to the analysis result and the service links included in the target service;

[0044] The feature extraction unit is used to sequentially capture the audio and video clips corresponding to each of the service links from the audio and video information, and perform feature extraction on the audio and video clips corresponding to each of the service links to obtain feature information of each of the service links.

[0045] Optionally, the user evaluation device may further include:

[0046] The determination module is used to determine the user's comprehensive evaluation result of the target service according to the user's evaluation result of each service link.

[0047] Optionally, the user evaluation device may further include:

[0048] The storage module is used to store the audio and video clips corresponding to each service link and the user's evaluation results of each service link in a database to update the evaluation model of each service link as training samples.

[0049] Optionally, the feature information may include but is not limited to at least one of the following:

[0050] Facial expression feature information, body movement feature information and speech semantic feature information.

[0051] In a third aspect, an embodiment of the present invention provides a user evaluation device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program can implement the steps of the above-mentioned user evaluation method when executed by the processor.

[0052] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program can implement the steps of the above-mentioned user evaluation method when executed by a processor.

[0053] In the embodiment of the present invention, by obtaining the audio and video information of the user during the target service process, according to the service links included in the target service, the audio and video clips corresponding to each service link in the audio and video information are sequentially extracted to obtain the feature information of each service link, and the feature information of each service link is sequentially analyzed to obtain the user's evaluation result of each service link. Compared with the prior art, the embodiment of the present invention can finely obtain the user's partial evaluation of a single link of the service, thereby fully understanding the user's evaluation of the service content, improving the accuracy of the evaluation result, and helping to optimize and improve the service quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments of the present invention will be 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 creative work.

[0055] Figure 1 is a flow chart of a user evaluation method according to an embodiment of the present invention;

[0056] Figure 2 is a flow chart of a user evaluation process according to an embodiment of the present invention;

[0057] Figure 3 It is a structural schematic diagram of a user evaluation system of a specific example of the present invention;

[0058] Figure 4 A schematic diagram of the structure of a user evaluation device according to an embodiment of the present invention;

[0059] Figure 5 Schematic diagram of the structure of a user evaluation device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0061] See also Figure 1 , Figure 1 1 is a flowchart of a user evaluation method provided by an embodiment of the present invention. The method is applied to a user evaluation device, which may be a user terminal or a server. Figure 1 As shown, the method may include the following steps:

[0062] Step 101: Acquire the user's audio and video information during the target service process.

[0063] In this embodiment, the target service may be any service provided to the user, such as a mobile phone purchase service, a card replacement service, a package replacement service, a mobile phone repair service, etc. provided by a communication operator. The audio and video information may be obtained by collecting the audio and video of the user at the service site in real time by an audio and video collection device (such as a camera and an audio collector) during the process of the user receiving the target service.

[0064] In one implementation, when a preset trigger condition of a target service is detected (for example, the target service is started), an audio and video acquisition device may be started to acquire audio and video of the user at the service site to obtain audio and video information of the user during the target service process.

[0065] Step 102: According to the service links included in the target service, feature extraction is performed on the audio and video clips corresponding to each service link in the audio and video information in turn to obtain feature information of each service link.

[0066] It is understandable that the service links included in the target service can be divided based on the service content of the target service, for example, it can include N, where N is an integer greater than or equal to 1. Taking the target service as the "machine purchase service" provided by the communication operator as an example, the service links included in this "machine purchase service" can be: experience model machine link, machine purchase work order filling link, payment link, signature link, etc.

[0067] In this step 102, when extracting features from the audio and video clips, for the video information in the audio and video clips, the frame images in the video information can be identified with the help of an existing image recognition algorithm (such as an image recognition model) to obtain corresponding feature information; and for the audio information in the audio and video clips, the audio information can be identified with the help of an existing audio recognition algorithm (such as an audio recognition model) to obtain corresponding feature information.

[0068] Optionally, the characteristic information of the above service link may include but is not limited to at least one of the following: facial expression characteristic information, body movement characteristic information, and speech semantic characteristic information, etc. Further, the facial expression characteristic information may include frowning, staring, anger, smiling, etc.; the body movement characteristic information may include shrugging, pinching waist, hugging arms, shaking head, nodding, etc.; the speech semantic characteristic information may include slow speech speed, fast speech speed, troublesome (semantic characteristics), etc.

[0069] Step 103: Analyze the characteristic information of each service link in turn to obtain the user's evaluation result of each service link.

[0070] In this embodiment, when analyzing the characteristic information of each service link, the analysis may be performed according to pre-set rules (such as the correspondence between pre-set characteristic information and evaluation results), or may be performed using a pre-trained evaluation model.

[0071] Optionally, the above step 103 may include:

[0072] The characteristic information of each service link is input into the pre-trained evaluation model of each service link in turn to obtain the user's evaluation result of each service link.

[0073] After obtaining the user's evaluation results for each service link with the help of the evaluation model, the audio and video clips corresponding to each service link and the user's evaluation results for each service link can be stored in the database as training samples to update the evaluation model of each service link, thereby ensuring the analysis accuracy of the evaluation model.

[0074] It should be noted that the above-mentioned pre-trained evaluation models correspond to service links, that is, each service link corresponds to a pre-trained evaluation model, and different service links correspond to different pre-trained evaluation models.

[0075] The user evaluation method of the embodiment of the present invention obtains the audio and video information of the user during the target service process, and extracts the features of the audio and video clips corresponding to each service link in the audio and video information in turn according to the service links included in the target service, obtains the feature information of each service link, and analyzes the feature information of each service link in turn, so as to obtain the user's evaluation result of each service link. Compared with the prior art, the embodiment of the present invention can obtain the user's partial evaluation of a single link of the service in detail, so as to fully understand the user's evaluation of the service content, improve the accuracy of the evaluation result, and help optimize and improve the service quality.

[0076] In order to obtain the user's evaluation results on the service link in real time, objectively and promptly, the audio and video information in the embodiment of the present invention can be obtained in real time during the target service process, and by monitoring the service link of the target service, the audio and video clips corresponding to the current service link can be used in real time to obtain the user's evaluation results on the current service link until the next service link, and the audio and video clips corresponding to the next service link can be used to obtain the user's evaluation results on the next service link until the service process ends.

[0077] When monitoring the service link of the target service, the monitoring can be achieved by analyzing the user's audio and video information, or by analyzing the service personnel's operation of the corresponding service system. Understandably, since the service content of different service links is generally different, for example, in the "machine purchase service", the machine purchase work order filling link requires the user to fill in the machine purchase work order, and the payment link requires the user to pay, so the service link can be monitored by analyzing the user's audio and video information during the service process.

[0078] Optionally, the above step 102 may include:

[0079] Analyze the audio and video information, and determine each service link of the target service in sequence according to the analysis result and the service links included in the target service;

[0080] The audio and video clips corresponding to each of the service links are sequentially intercepted from the audio and video information, and features are extracted from the audio and video clips corresponding to each of the service links to obtain feature information of each of the service links.

[0081] In this way, the service link that the user is in can be determined in real time, and with the help of feature extraction of corresponding audio and video clips, the user's evaluation results on the service link that he is in can be obtained in real time, objectively and promptly.

[0082] Furthermore, after obtaining the user's evaluation result on the service link, the evaluation result can be fed back to the current service personnel, so that the current service personnel can immediately self-check and adjust the service to the user according to the evaluation result.

[0083] In order to obtain the user's evaluation results of the corresponding service link by using the pre-trained evaluation model, the embodiment of the present invention can train the evaluation model before implementing the user evaluation. Optionally, the model training process in the embodiment of the present invention may include:

[0084] Get the service training collection;

[0085] Classifying the services in the service training set to obtain each category of services;

[0086] According to a preset rule, each type of service is divided into N service links, where N is an integer greater than or equal to 1;

[0087] For each service link of each type of service, the following processes are performed respectively:

[0088] Respectively obtaining a plurality of audio and video clips of a plurality of users corresponding to the service link, and a plurality of evaluation results of the plurality of users on the service link;

[0089] Extracting features from the multiple audio and video clips to obtain multiple feature information of the service link;

[0090] An evaluation model for the service link is trained by utilizing the multiple feature information of the service link and the multiple evaluation results.

[0091] The service training set may include multiple services for users. For example, taking the business hall service of a communication operator as an example, the corresponding service training set may be shown in Table 1 below:

[0092] Table 1

[0093] Service Training Collection Specific services Serves 1 Purchase service Serves 2 Card replacement service Serves 3 Change package service …… …… Service M Mobile phone repair service

[0094] Optionally, when the services in the service training set are classified, the services of the same category may be classified into one category to obtain each category of services. For example, the service training set shown in Table 1 may be classified into one category of services such as phone purchase service, card replacement service, package replacement service or phone repair service.

[0095] When dividing each type of service into service links, the division can be performed according to preset rules and specific service contents. For example, taking the "machine purchase service" in Table 1 above as an example, the service links obtained by dividing the "machine purchase service" can be shown in Table 2 below:

[0096] Table 2

[0097] Service link number Service content Session 1 Experience model machine Session 2 Filling out the purchase order Stage 3 Payment …… …… Link N Signature

[0098] It should be pointed out that each service link in the present embodiment corresponds to a pre-trained evaluation model. Therefore, when performing model training, a specific service link is targeted, that is, firstly, a training sample of the service link (including at least a plurality of audio and video clips and corresponding evaluation results) is obtained, and then the evaluation model of the service link is obtained by training with the training sample. The feature information obtained in the model training may also include facial expression feature information, body movement feature information and / or speech semantic feature information, etc. The specific model training algorithm can adopt an existing algorithm, such as a machine learning algorithm, a neural network deep learning algorithm, etc., and the embodiment of the present invention does not limit this.

[0099] For example, taking the “payment link” in Table 2 above as an example, the multiple audio and video clips of multiple users obtained may be shown in Table 3 below:

[0100] Table 3

[0101] Payment Audio and video clips User 1 Audio and video clip 1 User 2 Audio and video clip 2 User 3 Audio and video clip 3 …… …… User M Audio and video clips M

[0102] In at least one embodiment of the present invention, after the above step 103, the method may further include:

[0103] According to the evaluation results of the user on each of the service links, the comprehensive evaluation results of the user on the target service are determined.

[0104] The above evaluation results may be in the form of scores (e.g., 10-point system). When determining the comprehensive evaluation results of the corresponding service based on the evaluation results of the service links, the arithmetic mean or weighted mean may be used. In this way, the user's overall evaluation of the corresponding service can be understood.

[0105] For example, taking the “machine purchase service” in Table 1 above as an example, the evaluation results of each service link and the corresponding comprehensive evaluation results can be shown in Table 4 below:

[0106] Table 4

[0107]

[0108] Combine the following Figure 2 The user evaluation process of the embodiment of the present invention is described in detail.

[0109] In the embodiment of the present invention, taking the user accepting service A as an example, Figure 2 As shown, the corresponding user evaluation process may include the following steps:

[0110] Step 201: monitor whether service A is triggered;

[0111] Step 202: When it is detected that service A is triggered, the audio and video acquisition device is started to collect the audio and video information of the user during service A in real time;

[0112] Step 203: Analyze the collected audio and video information, determine the current service link of service A in turn, extract features of the audio and video clips corresponding to the current service link, analyze the corresponding feature information using the evaluation model of the current service link, obtain the user's evaluation result, and store the obtained evaluation result;

[0113] Step 204: monitor whether the current service node of service A is the last service node; if so, execute step 205; otherwise, continue to execute step 203;

[0114] Step 205: Analyze the corresponding feature information using the evaluation model of the last service link to obtain the user's evaluation result;

[0115] Step 206: When service A ends, the audio and video acquisition device is turned off to stop audio and video acquisition.

[0116] Step 207: Based on the evaluation results of the user on each link of service A, obtain the user's comprehensive evaluation result on service A.

[0117] It should be noted that in the specific example of the present invention, the user evaluation device may be a server (such as a backend server), and the corresponding user evaluation system may be as follows: Figure 3 As shown, it may include a user end 31 and a background service end 32 .

[0118] The user terminal 31 mainly includes an audio and video acquisition device 311 for acquiring audio and video information of the user during the service process. The audio and video acquisition device 311 is, for example, a camera and an audio collector.

[0119] The backend server 32 may include a service classification management module 321, a video processing and feature extraction module 322, a video feature training and modeling module 323, a video acquisition and control module 324, a video library module 325, a real-time evaluation module 326, a service category and service link identification module 327, an overall evaluation module 328 and other modules 329. Among them, the service classification management module 321 is mainly used for classification processing and storage of service classification and service links. The video processing and feature extraction module 322 is mainly used for preprocessing audio and video information and extracting feature information such as expression, action, and semantics. The video feature training and modeling module 323 is mainly used for feature training and modeling of audio and video samples according to preset rules. The video acquisition and control module is mainly used for collecting user audio and video information during the service process and controlling the audio and video acquisition device 311. The video library module 325 is mainly used for archiving and storing user audio and video information and evaluation results. The real-time evaluation module 326 is mainly used for real-time evaluation based on the feature information of the audio and video clips of the service link in which the user is currently located. The service category and service link identification module 327 is mainly used to detect and identify the service category and service link of the current user. The overall evaluation module 328 is mainly used to perform an overall evaluation of the current service according to preset rules. The other modules 329 can be selected as a customer service staff operation module, a real-time evaluation display module, etc.

[0120] It can be understood that the application scenarios of the embodiments of the present invention include but are not limited to traditional fields such as communications, finance, and catering, as well as emerging fields such as self-service vending and unmanned supermarkets, thereby providing automatic service evaluation for different fields and helping to improve their service and business quality.

[0121] See also Figure 4 , Figure 4 1 is a schematic diagram of the structure of a user evaluation device provided by an embodiment of the present invention. The user evaluation device is applied to a user evaluation device, which can be a user terminal or a server. Figure 4 As shown, the user evaluation device 40 includes:

[0122] The first acquisition module 41 is used to acquire the audio and video information of the user during the target service process;

[0123] A feature extraction module 42 is used to extract features of the audio and video clips corresponding to each service link in the audio and video information in turn according to the service links included in the target service, so as to obtain feature information of each service link;

[0124] The analysis module 43 is used to analyze the characteristic information of each service link in turn to obtain the user's evaluation result of each service link.

[0125] The user evaluation device of the embodiment of the present invention obtains the audio and video information of the user during the target service process, and extracts the features of the audio and video clips corresponding to each service link in the audio and video information in turn according to the service links included in the target service, obtains the feature information of each service link, and analyzes the feature information of each service link in turn, so as to obtain the user's evaluation result of each service link. Compared with the prior art, the embodiment of the present invention can obtain the user's partial evaluation of a single link of the service in a precise manner, so as to fully understand the user's evaluation of the service content and improve the accuracy of the evaluation result.

[0126] Optionally, the analysis module 43 is specifically used for:

[0127] The characteristic information of each service link is input into the pre-trained evaluation model of each service link in turn to obtain the user's evaluation result of each service link.

[0128] Optionally, the user evaluation device 40 may further include:

[0129] A second acquisition module is used to acquire a service training set;

[0130] A classification processing module, used for classifying the services in the service training set to obtain each type of service;

[0131] A division module, used to divide each type of service into N service links according to a preset rule, where N is an integer greater than or equal to 1;

[0132] The execution module is used to execute the following processes for each service link of each type of service:

[0133] Respectively obtaining a plurality of audio and video clips of a plurality of users corresponding to the service link, and a plurality of evaluation results of the plurality of users on the service link;

[0134] Extracting features from the multiple audio and video clips to obtain multiple feature information of the service link;

[0135] An evaluation model for the service link is trained by utilizing the multiple feature information of the service link and the multiple evaluation results.

[0136] Optionally, the feature extraction module 42 may include:

[0137] An analysis unit, configured to analyze the audio and video information, and determine each service link of the target service in sequence according to the analysis result and the service links included in the target service;

[0138] The feature extraction unit is used to sequentially capture the audio and video clips corresponding to each of the service links from the audio and video information, and perform feature extraction on the audio and video clips corresponding to each of the service links to obtain feature information of each of the service links.

[0139] Optionally, the user evaluation device 40 may further include:

[0140] The determination module is used to determine the user's comprehensive evaluation result of the target service according to the user's evaluation result of each service link.

[0141] Optionally, the user evaluation device 40 may further include:

[0142] The storage module is used to store the audio and video clips corresponding to each service link and the user's evaluation results of each service link in a database to update the evaluation model of each service link as training samples.

[0143] Optionally, the feature information may include but is not limited to at least one of the following:

[0144] Facial expression feature information, body movement feature information and speech semantic feature information.

[0145] In addition, an embodiment of the present invention also provides a user evaluation device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, can implement the various processes of the above-mentioned user evaluation method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described here.

[0146] For details, see Figure 5 As shown, the embodiment of the present invention further provides a user evaluation device, including a bus 51 , a transceiver 52 , an antenna 53 , a bus interface 54 , a processor 55 and a memory 56 .

[0147] In the embodiment of the present invention, the user evaluation device further includes: a computer program stored in the memory 56 and executable on the processor 55 .

[0148] Optionally, when the computer program is executed by the processor 55, the following steps may be implemented:

[0149] Obtain the user's audio and video information during the target service process;

[0150] According to the service links included in the target service, feature extraction is performed on the audio and video clips corresponding to each service link in the audio and video information in turn to obtain feature information of each service link;

[0151] The characteristic information of each service link is analyzed in turn to obtain the user's evaluation result of each service link.

[0152] It is understandable that in the embodiment of the present invention, when the computer program is executed by the processor 55, the above Figure 1 The various processes of the user evaluation method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0153] exist Figure 5 In the embodiment, the bus architecture (represented by bus 51) is shown, and bus 51 may include any number of interconnected buses and bridges, and bus 51 links various circuits including one or more processors represented by processor 55 and memory represented by memory 56. Bus 51 may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 54 provides an interface between bus 51 and transceiver 52. Transceiver 52 may be one element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on a transmission medium. Data processed by processor 55 is transmitted on a wireless medium via antenna 53, and further, antenna 53 also receives data and transmits the data to processor 55.

[0154] The processor 55 is responsible for managing the bus 51 and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management and other control functions. The memory 56 can be used to store data used by the processor 55 when performing operations.

[0155] Optionally, the processor 55 may be a CPU, an ASIC, an FPGA or a CPLD.

[0156] In addition, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which can implement the above-mentioned Figure 1 The various processes of the user evaluation method embodiment shown can achieve the same technical effect, and will not be described again here to avoid repetition.

[0157] Computer readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0158] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0159] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0160] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0161] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A user evaluation method, applied to a user evaluation device, characterized in that: include: Obtain the user's audio and video information during the target service process; According to the service links included in the target service, feature extraction is performed on the audio and video clips corresponding to each service link in the audio and video information in turn to obtain feature information of each service link; Analyze the characteristic information of each service link in turn to obtain the user's evaluation result of each service link; The characteristic information of each service link is analyzed in turn to obtain the user's evaluation result of each service link, including: Inputting the characteristic information of each service link into the pre-trained evaluation model of each service link in turn, and obtaining the user's evaluation result of each service link; Before obtaining the audio and video information of the user during the target service, the method further includes: Get the service training collection; Classifying the services in the service training set to obtain each category of services; According to a preset rule, each type of service is divided into N service links, where N is an integer greater than 1; For each service link of each type of service, the following processes are performed respectively: Respectively obtaining a plurality of audio and video clips of a plurality of users corresponding to the service link, and a plurality of evaluation results of the plurality of users on the service link; Extracting features from the multiple audio and video clips to obtain multiple feature information of the service link; An evaluation model for the service link is trained by utilizing the multiple feature information of the service link and the multiple evaluation results.

2. The method according to claim 1, characterized in that: According to the service links included in the target service, feature extraction is performed on the audio and video clips corresponding to each service link in the audio and video information in turn to obtain feature information of each service link, including: Analyze the audio and video information, and determine each service link of the target service in sequence according to the analysis result and the service links included in the target service; The audio and video clips corresponding to each of the service links are sequentially intercepted from the audio and video information, and features are extracted from the audio and video clips corresponding to each of the service links to obtain feature information of each of the service links.

3. The method according to claim 1, characterized in that After obtaining the user's evaluation result for each of the service links, the method further includes: According to the evaluation results of the user on each of the service links, the comprehensive evaluation results of the user on the target service are determined.

4. The method according to claim 1, characterized in that: After obtaining the user's evaluation result for each of the service links, the method further includes: The audio and video clips corresponding to each service link and the user's evaluation results for each service link are stored in the database to serve as training samples to update the evaluation model of each service link.

5. The method according to any one of claims 1 to 4, characterized in that The characteristic information includes at least one of the following: Facial expression feature information, body movement feature information and speech semantic feature information.

6. A user evaluation device, applied to a user evaluation device, characterized in that: include: The first acquisition module is used to obtain the audio and video information of the user during the target service process; A feature extraction module, configured to extract features from the audio and video clips corresponding to each service link in the audio and video information in turn according to the service links included in the target service, so as to obtain feature information of each service link; An analysis module, used to analyze the characteristic information of each service link in turn to obtain the user's evaluation result of each service link; The analysis module is specifically used to: sequentially input the characteristic information of each service link into the pre-trained evaluation model of each service link to obtain the user's evaluation result of each service link; The user evaluation device also includes: A second acquisition module is used to acquire a service training set; A classification processing module, used for classifying the services in the service training set to obtain each type of service; A division module, used to divide each type of service into N service links according to a preset rule, where N is an integer greater than; The execution module is used to execute the following processes for each service link of each type of service: Respectively obtaining a plurality of audio and video clips of a plurality of users corresponding to the service link, and a plurality of evaluation results of the plurality of users on the service link; Extracting features from the multiple audio and video clips to obtain multiple feature information of the service link; An evaluation model for the service link is trained by utilizing the multiple feature information of the service link and the multiple evaluation results.

7. A user evaluation device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by the processor, the steps of the user evaluation method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the user evaluation method according to any one of claims 1 to 5 are implemented.

Images