WO2023236730A1 - 音视频处理性能测试方法及装置 - Google Patents

音视频处理性能测试方法及装置 Download PDF

Info

Publication number
WO2023236730A1
WO2023236730A1 PCT/CN2023/094308 CN2023094308W WO2023236730A1 WO 2023236730 A1 WO2023236730 A1 WO 2023236730A1 CN 2023094308 W CN2023094308 W CN 2023094308W WO 2023236730 A1 WO2023236730 A1 WO 2023236730A1
Authority
WO
WIPO (PCT)
Prior art keywords
audio
video
video data
information
neural network
Prior art date
Application number
PCT/CN2023/094308
Other languages
English (en)
French (fr)
Inventor
蔡禄汀
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2023236730A1 publication Critical patent/WO2023236730A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/044Recurrent networks, e.g. Hopfield networks
    • G06N3/0442Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/04Architecture, e.g. interconnection topology
    • G06N3/0464Convolutional networks [CNN, ConvNet]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00Computing arrangements based on biological models
    • G06N3/02Neural networks
    • G06N3/08Learning methods
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04MTELEPHONIC COMMUNICATION
    • H04M3/00Automatic or semi-automatic exchanges
    • H04M3/22Arrangements for supervision, monitoring or testing
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04MTELEPHONIC COMMUNICATION
    • H04M3/00Automatic or semi-automatic exchanges
    • H04M3/42Systems providing special services or facilities to subscribers
    • H04M3/50Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing

Definitions

  • the embodiments of the present disclosure relate to the field of communications, and specifically, to an audio and video processing performance testing method and device.
  • Embodiments of the present disclosure provide an audio and video processing performance testing method and device to at least solve the problem in related technologies that the audio and video processing quality and performance of a call center cannot be efficiently and accurately evaluated and tested.
  • a computer-readable storage medium is also provided.
  • a computer program is stored in the computer-readable storage medium, wherein the computer program is configured to execute any of the above methods when running. Steps in Examples.
  • an electronic device including a memory and a processor.
  • a computer program is stored in the memory, and the processor is configured to run the computer program to perform any of the above. Steps in method embodiments.
  • Figure 1 is a hardware structure block diagram of a mobile terminal according to an audio and video processing performance testing method according to an embodiment of the present disclosure
  • Figure 2 is a schematic diagram of the network architecture of the audio and video processing performance testing method according to an embodiment of the present disclosure
  • Figure 4 is a flow chart of an audio and video processing performance testing method according to an embodiment of the present disclosure
  • Figure 5 is a structural block diagram of an audio and video processing performance testing device according to an embodiment of the present disclosure
  • Figure 6 is a block diagram of the data processing module of the audio and video processing performance testing device according to an embodiment of the present disclosure
  • Figure 7 is a structural block diagram of an audio and video processing performance testing device according to an embodiment of the present disclosure.
  • Figure 8 is a flow chart of an audio and video processing performance testing method according to an embodiment of the disclosed scenario.
  • FIG. 1 is a hardware structure block diagram of a mobile terminal of the audio and video processing performance testing method according to an embodiment of the present disclosure.
  • the mobile terminal may include one or more (only one is shown in Figure 1) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the above-mentioned mobile terminal may also include a transmission device 106 and an input and output device 108 for communication functions.
  • processors 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA
  • a memory 104 for storing data
  • the above-mentioned mobile terminal may also include a transmission device 106 and an input and output device 108 for communication functions.
  • the structure shown in Figure 1 is only illustrative, and it does not limit the structure of the above-mentioned mobile terminal.
  • the mobile terminal may also include more or fewer components than shown in FIG. 1 , or have a different configuration than shown in FIG. 1 .
  • the memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the audio and video processing performance testing method in the embodiment of the present disclosure.
  • the processor 102 runs the computer program stored in the memory 104, Thereby executing various functional applications and data processing, that is, realizing the above method.
  • Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory.
  • the memory 104 may further include memory located remotely relative to the processor 102, and these remote memories may be connected to the mobile terminal through a network. Examples of the above-mentioned networks include but are not limited to the Internet, intranets, local area networks, mobile communication networks and combinations thereof.
  • Transmission device 106 is used to receive or send data via a network.
  • Specific examples of the above-mentioned network may include a wireless network provided by a communication provider of the mobile terminal.
  • the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station to communicate with the Internet.
  • the transmission device 106 may be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet wirelessly.
  • RF Radio Frequency
  • the audio and video processing performance test network architecture 20 includes: a call center access module 210, an audio and video analysis module 220, and a configuration module 230.
  • the call center access module is used to access different call centers that need to be tested through the SIP protocol;
  • the audio and video analysis module is used to integrate the CNN+LSTM neural network, which can use audio and video data to train the neural network for call analysis. Audio and video streams in the process;
  • the configuration module is used to configure models and parameters related to performance testing.
  • FIG. 3 is a flow chart of the audio and video processing performance testing method according to an embodiment of the present disclosure, as shown in Figure 3. The process includes the following steps:
  • Step S302 construct an audio and video data set and mark the audio and video data set
  • Step S304 input the marked audio and video data set into the convolutional neural network cascade long short-term memory neural network CNN+LSTM for training;
  • Step S306 configure the parameter information of the call model, and set the quality threshold of audio and video data and the system resource usage information threshold;
  • Step S308 input audio and video data from the terminal into the trained CNN+LSTM neural network model to obtain audio and video quality information;
  • the labeled audio and video data set is input into CNN+LSTM for training; and then the audio and video data from the terminal is input into the trained CNN+LSTM neural network model to obtain audio and video quality information. ; Determine whether the audio and video quality information and the preconfigured system resource usage information satisfy the requirement of being less than or equal to the corresponding threshold. If not, increase the number of terminals and repeatedly input the audio and video data from the terminals to all terminals.
  • the technology cannot accurately evaluate and detect audio and video processing quality and performance issues in the call center, eliminating a large amount of manual participation and improving testing efficiency.
  • the execution subject of the above steps may be a base station, a terminal, or other processor platform that can be equipped with the programs required to execute the above method, but is not limited thereto.
  • each of the degradation methods is divided into multiple levels and a degradation score vector is constructed, including at least one of the following: non-blurred video, 3*3, 5*5, 9 *9, 15*15 convolution kernels of five sizes are used for Gaussian filtering; video frame extraction is performed at intervals of 5 frames, 10 frames, 15 frames, 30 frames, without frame extraction; audio and video out-of-synchronization is processed according to -5s, -2s, 0s, 2s, and 5s are processed asynchronously; audio noise is added according to the mean value of zero, four increasing variances add Gaussian white noise and no noise processing; a degradation score vector is constructed, where a1 represents the video blur level score, and a2 represents The grade score of video frame extraction, a3 represents the grade score of audio and video out of synchronization, a4 represents the grade score of audio plus noise, and the highest score, that is, the undegraded video quality score vector is [5, 5, 5, 5].
  • the parameter information of the call model includes at least one of the following: number of call channels; transmission time of audio and video data; video resolution; encoding method of audio and video data; transmission bit rate of audio and video data .
  • the method further includes: configuring access information of a target call center, where the target call center supports multiple different call models; the terminal Access the target call center through the SIP protocol.
  • Figure 4 is a flow chart of an audio and video processing performance testing method according to an embodiment of the present disclosure. As shown in Figure 4, the process includes the following steps:
  • Step S406 configure the parameter information of the call model, and set the quality threshold of audio and video data and the system resource usage information threshold;
  • Step S408 Configure access information of the target call center, where the target call center supports multiple different call models
  • Step S412 input audio and video data from the terminal into the trained CNN+LSTM neural network model to obtain audio and video quality information;
  • the method according to the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is Better implementation.
  • the technical solutions of the embodiments of the present disclosure can be embodied in the form of software products in essence or those that contribute to the existing technology.
  • the computer software products are stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), including several instructions to cause a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in various embodiments of the embodiments of the present disclosure.
  • module, unit may be a combination of software and/or hardware that implements a predetermined function.
  • the apparatus described in the following embodiments is preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.
  • FIG. 5 is a structural block diagram of an audio and video processing performance testing device according to an embodiment of the present disclosure. As shown in Figure 5, the audio and video processing performance test device Frequency processing performance testing device 50, including:
  • the data processing module 510 is configured to construct an audio and video data set and mark the audio and video data set;
  • the neural network training module 520 is configured to input the labeled audio and video data set to the convolutional neural network cascade long short-term memory neural network CNN+LSTM for training;
  • the first configuration module 530 is configured to configure the parameter information of the call model, and set the quality threshold of audio and video data and the system resource usage information threshold;
  • the testing module 540 is configured to input audio and video data from the terminal into the trained CNN+LSTM neural network model to obtain audio and video quality information;
  • the judgment loop module 550 is configured to judge whether the audio and video quality information and the preconfigured system resource usage information meet the requirements of being less than or equal to the corresponding threshold. If not, increase the number of terminals and return to the test module 540 , until the audio and video quality information and the preconfigured system resource usage information are less than or equal to the corresponding threshold;
  • the reporting module 560 is configured to output processing performance test results based on the audio and video quality information and the preconfigured system resource usage information when the audio and video quality information and the preconfigured system resource usage information are less than or equal to the corresponding thresholds.
  • FIG. 6 is a structural block diagram of a data processing module of an audio and video processing performance testing device according to an embodiment of the present disclosure.
  • the data processing module 510 further includes at least one of the following: degradation unit 610 , is configured to degrade the audio and video data sets according to different degradation methods; the grade classification unit 620 is configured to divide each of the degradation methods into multiple grades, and construct a degradation score vector; the marking unit 630 is configured to classify the The audio and video data set is marked according to the degradation method and classification level.
  • different degradation methods in the degradation unit 610 include at least one of the following: video blur; video frame extraction; audio and video desynchronization; and audio noise addition.
  • the parameter information of the call model configured by the first configuration module 530 includes at least one of the following: number of call channels; transmission time of audio and video data; video resolution; encoding method of audio and video data; audio and video Data transmission code rate.
  • FIG. 7 is a structural block diagram of an audio and video processing performance testing device according to an embodiment of the present disclosure.
  • the audio and video processing performance testing device 70 except the various modules mentioned in FIG. 5
  • it also includes: a second configuration module 710, configured to configure the access information of the target call center, wherein the target call center supports multiple different call models; a terminal access module 720, configured to configure the terminal Access the target call center through the SIP protocol.
  • each of the above modules and units can be implemented through software or hardware.
  • it can be implemented in the following ways, but is not limited to this: the above modules and units are all located in the same processor; or, each of the above Modules and units are located in different processors in any combination.
  • Embodiments of the present disclosure also provide a computer-readable storage medium that stores a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.
  • the above-mentioned computer-readable storage media may include but is not limited to: U disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), mobile hard disk, Various media such as magnetic disks or optical disks that can store computer programs.
  • An embodiment of the present disclosure also provides an electronic device, including a memory and a processor.
  • a computer program is stored in the memory, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.
  • the above-mentioned electronic device may further include a transmission device and an input and output device, wherein, The transmission device is connected to the above-mentioned processor, and the input and output device is connected to the above-mentioned processor.
  • CNN+LSTM Convolutional Neural Network Cascade Long Short-term Memory Neural Network
  • the basic idea is to use convolutional neural networks to extract audio and video features, and use LSTM To generate descriptions, it is one of the few joint applications of the two popular models in deep learning.
  • the call center audio and video processing performance detection system based on deep learning provided by the embodiment of the present disclosure utilizes the capabilities of CNN+LSTM neural network to eliminate a large amount of manual participation and efficiently and accurately evaluate and detect the audio and video processing quality and performance of the call center.
  • FIG. 8 is a flow chart of an audio and video processing performance testing method according to an embodiment of the disclosed scenario. As shown in Figure 8, the process includes the following steps:
  • Step S802 train CNN+LSTM neural network
  • video blur is unblurred, 3*3, 5*5, 9*9, 15*15 convolution kernels of five sizes are used for Gaussian filtering; video frame extraction is based on intervals of 5 frames, 10 frames, 15 frames, 30 frames, and no frame extraction; audio and video asynchronous processing is based on -5s, -2s, 0s, 2s , 5s for asynchronous processing; the audio noise is added according to the mean value of zero, Gaussian white noise is added with four increasing variances, and no noise is added. Therefore, the degradation score of the final training set can form a four-dimensional vector, in which the highest score, that is, the undegraded video quality score vector is [5, 5, 5, 5].
  • the training set is processed in 4 ways (video blur, video frame extraction, audio and video asynchronous processing, and audio noise addition) and 5 levels of permutation and combination processing to obtain different Score markers. , send it to the constructed CNN+LSTM neural network for training to obtain a pre-trained neural network;
  • Step S804 configure model parameters
  • Configure the call model information of the test system that is, how many members are required for a call, the time for each member to send audio and video streams, video resolution, audio and video encoding method, transmission bit rate and other information.
  • Step S806 simulate the terminal to access the call center
  • Video stream (can also be an actual audio and video stream);
  • Step S808 simulate the terminal to receive audio and video streams
  • Step S810 CNN+LSTM neural network performs audio and video quality analysis
  • CNN+LSTM neural network to obtain audio and video quality information
  • test system can monitor the system resource (CPU, memory) usage information of the server where the call center is located;
  • Step S812 determine whether the threshold is reached
  • step S810 it is judged whether it has dropped to the threshold configured in step S804. If it has not reached the threshold below, increase the number of call channels (that is, increase the number of analog terminals) and increase the video processing pressure. And jump to step S808, otherwise, proceed to step S814;
  • Step S814 output a performance test result report
  • a performance test result report is finally output, and the process ends.
  • the audio and video processing performance testing method and device provided by the embodiments of the present disclosure can be applied to call center audio and video performance quality testing.
  • the call center needs to have normal network communication conditions, have audio and video processing capabilities, and be able to conduct terminal operations through the SIP standard protocol. Registration access and media negotiation.
  • the purpose of the audio and video processing performance testing method and device provided by the embodiments of the present disclosure is to provide a performance detection system for the audio and video processing capabilities of a new generation call center, combined with deep learning and other technologies, to detect the deficiencies and limitations of the audio and video processing of the call center. Bottlenecks to help improve call center service quality.
  • Embodiments of the present disclosure provide audio and video processing capability detection tools for each call center system, and can provide a call center audio and video processing quality and performance detection system based on deep learning. Provide reliable performance detection indicators for the call center's audio and video processing capabilities, help the call center discover performance bottlenecks, improve problems, and enhance audio and video processing capabilities and user experience.
  • each module, unit or step of the above-mentioned embodiments of the present disclosure can be implemented by a general computing device, and they can be concentrated on a single computing device, or distributed among multiple computing devices. On the network formed, they may be implemented in program code executable by a computing device, so that they may be stored in a storage device and executed by the computing device, and in some cases, may be in a sequence different from that described here.
  • the steps shown or described are performed either individually as individual integrated circuit modules, or as multiple modules or steps among them as a single integrated circuit module. As such, disclosed embodiments are not limited to any specific combination of hardware and software.

Abstract

本公开实施例提供了一种音视频处理性能测试方法及装置,通过本公开实施例,将标记后的音视频数据集输入至CNN+LSTM进行训练;再将来自终端的音视频数据输入至训练好的CNN+LSTM神经网络模型,获得音视频质量信息;判断音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加终端的数量,并重复执行将来自终端的音视频数据输入至训练好的CNN+LSTM神经网络模型,获得音视频质量信息的步骤,直至音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值,输出处理性能测试结果。

Description

音视频处理性能测试方法及装置
相关申请的交叉引用
本申请基于2022年6月6日提交的发明名称为“音视频处理性能测试方法及装置”的中国专利申请CN202210633014.1,并且要求该专利申请的优先权,通过引用将其所公开的内容全部并入本申请。
技术领域
本公开实施例涉及通信领域,具体而言,涉及一种音视频处理性能测试方法及装置。
背景技术
随着5G,音视频等技术的发展,为客户提供更好的体验和更高的服务质量,呼叫中心已经从简单的呼叫接听处理发展成提供强大音视频交互能力的新一代系统,因此,如何检测并评估呼叫中心的音视频处理能力,从而提升呼叫中心的服务质量是新一代呼叫中心亟待解决的问题。
当前呼叫中心的性能检测工具主要关注于同时坐席在线数,实时呼叫数,最大呼叫数等指标,对于音视频质量的评估大多仅限于音视频流抓包,通过专业人员人工分析结合坐席用户评价,这种性能检测评价方式极为低效。
发明内容
本公开实施例提供了一种音视频处理性能测试方法及装置,以至少解决相关技术中无法高效准确的评估检测呼叫中心的音视频处理质量和性能的问题。
根据本公开的一个实施例,提供了一种音视频处理性能测试方法,包括:构建音视频数据集,并对所述音视频数据集进行标记;将标记后的所述音视频数据集输入至卷积神经网络级联长短期记忆神经网络CNN+LSTM进行训练;配置呼叫模型的参数信息,并设置音视频数据的质量阈值和系统资源使用信息阈值;将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息;判断所述音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加所述终端的数量,并重复执行将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息的步骤,直至所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值,输出处理性能测试结果。
根据本公开的另一实施例,还提供了一种音视频处理性能测试装置,包括:数据处理模块,设置为构建音视频数据集,并对所述音视频数据集进行标记;神经网络训练模块,设置为将标记后的所述音视频数据集输入至卷积神经网络级联长短期记忆神经网络CNN+LSTM进行训练;第一配置模块,设置为配置呼叫模型的参数信息,并设置音视频数据的质量阈值和系统资源使用信息阈值;测试模块,设置为将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息;判断循环模块,设置为判断所述音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加 所述终端的数量,返回至所述测试模块,直至所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值;报告模块,设置为在所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值时,根据所述音视频质量信息和系统资源使用信息输出处理性能测试结果。
根据本公开的又一个实施例,还提供了一种计算机可读存储介质,所述计算机可读存储介质中存储有计算机程序,其中,所述计算机程序被设置为运行时执行上述任一项方法实施例中的步骤。
根据本公开的又一个实施例,还提供了一种电子装置,包括存储器和处理器,所述存储器中存储有计算机程序,所述处理器被设置为运行所述计算机程序以执行上述任一项方法实施例中的步骤。
附图说明
图1是根据本公开实施例的音视频处理性能测试方法的移动终端的硬件结构框图;
图2是根据本公开实施例的音视频处理性能测试方法的网络架构的示意图;
图3是根据本公开实施例的音视频处理性能测试方法的流程图;
图4是根据本公开实施例的音视频处理性能测试方法的流程图;
图5是根据本公开实施例的音视频处理性能测试装置的结构框图;
图6是根据本公开实施例的音视频处理性能测试装置的数据处理模块结构框图;
图7是根据本公开实施例的音视频处理性能测试装置的结构框图;
图8是根据本公开场景实施例的音视频处理性能测试方法的流程图。
具体实施方式
下文中将参考附图并结合实施例来详细说明本公开实施例。
需要说明的是,本公开的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。
本公开实施例中所提供的方法实施例可以在移动终端、计算机终端或者类似的运算装置中执行。以运行在移动终端上为例,图1是本公开实施例的音视频处理性能测试方法的移动终端的硬件结构框图。如图1所示,移动终端可以包括一个或多个(图1中仅示出一个)处理器102(处理器102可以包括但不限于微处理器MCU或可编程逻辑器件FPGA等的处理装置)和用于存储数据的存储器104,其中,上述移动终端还可以包括用于通信功能的传输设备106以及输入输出设备108。本领域普通技术人员可以理解,图1所示的结构仅为示意,其并不对上述移动终端的结构造成限定。例如,移动终端还可包括比图1中所示更多或者更少的组件,或者具有与图1所示不同的配置。
存储器104可用于存储计算机程序,例如,应用软件的软件程序以及模块,如本公开实施例中的音视频处理性能测试方法对应的计算机程序,处理器102通过运行存储在存储器104内的计算机程序,从而执行各种功能应用以及数据处理,即实现上述的方法。存储器104可包括高速随机存储器,还可包括非易失性存储器,如一个或者多个磁性存储装置、闪存、或者其他非易失性固态存储器。在一些实例中,存储器104可进一步包括相对于处理器102远程设置的存储器,这些远程存储器可以通过网络连接至移动终端。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
传输设备106用于经由一个网络接收或者发送数据。上述的网络具体实例可包括移动终端的通信供应商提供的无线网络。在一个实例中,传输设备106包括一个网络适配器(Network Interface Controller,NIC),其可通过基站与其他网络设备相连从而可与互联网进行通讯。在一个实例中,传输设备106可以为射频(Radio Frequency,RF)模块,其用于通过无线方式与互联网进行通讯。
本公开实施例可以运行于图2所示的网络架构上,如图2所示,该音视频处理性能测试网络架构20包括:呼叫中心接入模块210、音视频分析模块220和配置模块230。其中,呼叫中心接入模块用于通过SIP协议用于接入不同的需要测试的呼叫中心;音视频分析模块用于集成CNN+LSTM神经网络,可以利用音视频数据训练神经网络,用于分析呼叫流程中的音视频流;配置模块用于配置性能测试相关的模型和参数。
在本实施例中提供了一种运行于上述移动终端或网络架构的音视频处理性能测试方法,图3是根据本公开实施例的音视频处理性能测试方法的流程图,如图3所示,该流程包括如下步骤:
步骤S302,构建音视频数据集,并对所述音视频数据集进行标记;
步骤S304,将标记后的所述音视频数据集输入至卷积神经网络级联长短期记忆神经网络CNN+LSTM进行训练;
步骤S306,配置呼叫模型的参数信息,并设置音视频数据的质量阈值和系统资源使用信息阈值;
步骤S308,将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息;
步骤S310,判断所述音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加所述终端的数量,并重复执行步骤S308,直至所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值,输出处理性能测试结果。
通过上述步骤,通过将标记后的所述音视频数据集输入至CNN+LSTM进行训练;再将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息;判断所述音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加所述终端的数量,并重复执行将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息的步骤,直至所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值,输出处理性能测试结果,解决了相关技术中无法准确的评估检测呼叫中心的音视频处理质量和性能的问题,达到排除大量人工参与,提高测试效率的效果。
其中,上述步骤的执行主体可以为基站、终端等其他可以搭载上述方法执行所需程序的处理器平台上,但不限于此。
在一个示例性实施例中,所述对音视频数据集进行标记,包括以下至少之一:将所述音视频数据集按不同劣化方式进行劣化;将每种所述劣化方式划分为多个等级,构建劣化评分向量;根据所述劣化方式和划分等级对所述音视频数据集进行标记。
在一个示例性实施例中,所述不同劣化方式至少包括以下之一:视频模糊;视频抽帧; 音视频不同步;音频加噪。
在一个示例性实施例中,所述将每种所述劣化方式划分为多个等级,构建劣化评分向量,包括以下至少之一:视频模糊采用不模糊,采用3*3,5*5,9*9,15*15五种大小的卷积核进行高斯滤波;视频抽帧按照间隔5帧,10帧,15帧,30帧,不抽帧处理;音视频不同步按照-5s,-2s,0s,2s,5s进行不同步处理;音频加噪按照均值为零,四个递增的方差添加高斯白噪声以及不加噪处理;构建劣化评分向量,其中,a1表示视频模糊的等级分数,a2表示视频抽帧的等级分数,a3表示音视频不同步的等级分数,a4表示音频加噪的等级分数,,最高分即未劣化的视频质量得分向量为[5,5,5,5]。
在一个示例性实施例中,所述呼叫模型的参数信息包括以下至少之一:呼叫路数;音视频数据的发送时间;视频分辨率;音视频数据的编码方式;音视频数据的传输码率。
在一个示例性实施例中,在所述配置呼叫模型的参数信息之后,还包括:配置目标呼叫中心的接入信息,其中所述目标呼叫中心支持多个不同的所述呼叫模型;所述终端通过SIP协议接入所述目标呼叫中心。图4是根据本公开实施例的音视频处理性能测试方法的流程图,如图4所示,该流程包括如下步骤:
步骤S402,构建音视频数据集,并对所述音视频数据集进行标记;
步骤S404,将标记后的所述音视频数据集输入至卷积神经网络级联长短期记忆神经网络CNN+LSTM进行训练;
步骤S406,配置呼叫模型的参数信息,并设置音视频数据的质量阈值和系统资源使用信息阈值;
步骤S408,配置目标呼叫中心的接入信息,其中所述目标呼叫中心支持多个不同的所述呼叫模型;
步骤S410,所述终端通过SIP协议接入所述目标呼叫中心;
步骤S412,将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息;
步骤S414,判断所述音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加所述终端的数量,并重复执行步骤S412,直至所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值,输出处理性能测试结果。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到根据上述实施例的方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本公开实施例的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台终端设备(可以是手机,计算机,服务器,或者网络设备等)执行本公开实施例的各个实施例所述的方法。
在本实施例中还提供了一种音视频处理性能测试装置,该装置用于实现上述实施例及优选实施方式,已经进行过说明的不再赘述。如以下所使用的,术语“模块、单元”可以实现预定功能的软件和/或硬件的组合。尽管以下实施例所描述的装置较佳地以软件来实现,但是硬件,或者软件和硬件的组合的实现也是可能并被构想的。
图5是根据本公开实施例的音视频处理性能测试装置的结构框图,如图5所示,该音视 频处理性能测试装置50,包括:
数据处理模块510,设置为构建音视频数据集,并对所述音视频数据集进行标记;
神经网络训练模块520,设置为将标记后的所述音视频数据集输入至卷积神经网络级联长短期记忆神经网络CNN+LSTM进行训练;
第一配置模块530,设置为配置呼叫模型的参数信息,并设置音视频数据的质量阈值和系统资源使用信息阈值;
测试模块540,设置为将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息;
判断循环模块550,设置为判断所述音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加所述终端的数量,返回至所述测试模块540,直至所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值;
报告模块560,设置为在所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值时,根据所述音视频质量信息和系统资源使用信息输出处理性能测试结果。
在一个示例性实施例中,图6是根据本公开实施例的音视频处理性能测试装置的数据处理模块结构框图,如图6所示,数据处理模块510进一步包括以下至少之一:劣化单元610,设置为将所述音视频数据集按不同劣化方式进行劣化;等级划分单元620,设置为将每种所述劣化方式划分为多个等级,构建劣化评分向量;标记单元630,设置为根据所述劣化方式和划分等级对所述音视频数据集进行标记。
在一个示例性实施例中,劣化单元610中的不同劣化方式至少包括以下之一:视频模糊;视频抽帧;音视频不同步;音频加噪。
在一个示例性实施例中,第一配置模块530配置的呼叫模型的参数信息包括以下至少之一:呼叫路数;音视频数据的发送时间;视频分辨率;音视频数据的编码方式;音视频数据的传输码率。
在一个示例性实施例中,图7是根据本公开实施例的音视频处理性能测试装置的结构框图,如图7所示,该音视频处理性能测试装置70除了图5中提到的各个模块外,还包括:第二配置模块710,设置为配置目标呼叫中心的接入信息,其中所述目标呼叫中心支持多个不同的所述呼叫模型;终端接入模块720,设置为将所述终端通过SIP协议接入所述目标呼叫中心。
需要说明的是,上述各个模块、单元是可以通过软件或硬件来实现的,对于后者,可以通过以下方式实现,但不限于此:上述模块、单元均位于同一处理器中;或者,上述各个模块、单元以任意组合的形式分别位于不同的处理器中。
本公开实施例还提供了一种计算机可读存储介质,该计算机可读存储介质中存储有计算机程序,其中,该计算机程序被设置为运行时执行上述任一项方法实施例中的步骤。
在一个示例性实施例中,上述计算机可读存储介质可以包括但不限于:U盘、只读存储器(Read-Only Memory,ROM)、随机存取存储器(Random Access Memory,RAM)、移动硬盘、磁碟或者光盘等各种可以存储计算机程序的介质。
本公开实施例还提供了一种电子装置,包括存储器和处理器,该存储器中存储有计算机程序,该处理器被设置为运行计算机程序以执行上述任一项方法实施例中的步骤。
在一个示例性实施例中,上述电子装置还可以包括传输设备以及输入输出设备,其中, 该传输设备和上述处理器连接,该输入输出设备和上述处理器连接。
本实施例中的具体示例可以参考上述实施例及示例性实施方式中所描述的示例,本实施例在此不再赘述。
为了使得本领域的普通技术人员能够更好地理解本公开实施例的技术方案,下面将结合具体的场景实施例对本公开实施例的技术方案进行阐述。
场景实施例一
当前深度学习技术高速发展,利用深度学习中的神经网络技术可以训练出实现多种功能的神经网络。其中,CNN+LSTM(卷积神经网络级联连长短期记忆神经网络)对处理音视频等时间序列具有广泛的应用,其基本思想是利用卷积神经网络来做音视频的特征提取,利用LSTM来生成描述,是深度学习中热门的两大模型为数不多的联合应用。本公开实施例提供的基于深度学习的呼叫中心音视频处理性能检测系统,利用CNN+LSTM神经网络的能力,能够排除人工的大量参与,高效准确的评估检测呼叫中心的音视频处理质量和性能。
图8是根据本公开场景实施例的音视频处理性能测试方法的流程图,如图8所示,该流程包括以下步骤:
步骤S802,训练CNN+LSTM神经网络;
首先利用网络获取大量音视频构建数据集,分类形成多个标准分辨率(720p,1080p,2k,4k)的训练集和测试集,通过将训练集的原始音视频按照不同方式进行劣化,劣化方式分为视频模糊,视频抽帧,音视频不同步处理,音频加噪,对每种劣化按强度分为5个等级,其中视频模糊采用不模糊,3*3,5*5,9*9,15*15五种大小的卷积核进行高斯滤波;视频抽帧按照间隔5帧,10帧,15帧,30帧,不抽帧;音视频不同步处理按照-5s,-2s,0s,2s,5s进行不同步处理;音频加噪按照均值为零,四个递增的方差添加高斯白噪声以及不加噪处理。因此,最终训练集的劣化打分可构成一个四维向量,其中,最高分即未劣化的视频质量得分向量为[5,5,5,5]。
利用上述的多个劣化维度的Score为目标,将训练集进行4种方式(视频模糊,视频抽帧,音视频不同步处理,音频加噪),5个等级的排列组合处理得到不同的Score标记,送入构建的CNN+LSTM神经网络进行训练得到预训练好的神经网络;
步骤S804,配置模型参数;
配置测试系统的呼叫模型信息,即一通呼叫需要多少个成员,每个成员发送音视频流的时间,视频的分辨率,音视频的编码方式,传输码率等信息。配置初始呼叫路数。配置音视频质量阈值Score,根据具体联络中心对视频质量的要求,从四个维度进行打分,例如[4,4,4,4],即视频质量最低需要达到步骤S802所述的四个劣化强度的4分;并配置呼叫中心所在服务端的系统资源(CPU,内存)使用信息及其阈值;
步骤S806,模拟终端接入呼叫中心;
配置目标呼叫中心的接入信息,通过SIP协议接入目标呼叫中心,根据步骤S804中配置的模型以及呼叫的路数,模拟终端(提供音视频流),进行呼叫,发送步骤S802中测试集音视频流(也可以是实际的音视频流);
步骤S808,模拟终端接收音视频流;
将各个模拟终端接收到的音视频流送入步骤S802中训练好的CNN+LSTM神经网络;
步骤S810,CNN+LSTM神经网络进行音视频质量分析;
通过CNN+LSTM神经网络进行音视频质量分析,得到音视频质量信息,和测试系统可监控呼叫中心所在服务端的系统资源(CPU,内存)使用信息;
步骤S812,判断是否达到阈值;
根据步骤S810中的质量和性能信息和CPU使用信息判断是否下降到步骤S804中配置的阈值,以下,若未到达,则增加呼叫路数(即增加模拟终端的数量),加大视频处理压力,并跳转到步骤S808,否则,进行步骤S814;
步骤S814,输出性能测试结果报告;
根据步骤S810得到的参数信息,最终输出性能测试结果报告,结束流程。
其中,本领域的普通技术人员应该知道,本场景实施例中的等级划分的等级数、划分方式、劣化方式、评分方式都是举例说明,并不限制于仅此一种方式。
本公开实施例提供的音视频处理性能测试方法及装置可以应用于呼叫中心音视频性能质量检测,其中,呼叫中心需要具备正常的网络通讯条件,具备音视频处理能力,能通过SIP标准协议进行终端的注册接入以及媒体协商。
本公开实施例提供的音视频处理性能测试方法及装置,目的在于提供一种针对新一代呼叫中心音视频处理能力的性能检测系统,结合深度学习等技术,检测呼叫中心的音视频处理的不足与瓶颈,为提高呼叫中心服务质量提供帮助。
本公开实施例利用丰富的互联网音视频资源构建数据集,分类形成多个分辨率的训练集和测试集,通过将训练集的原始音视频按照不同等级进行劣化并标记,训练测试系统的CNN+LSTM神经网络识别视频质量的能力。测试系统提供自定义音视频呼叫模型能力配置参数,通过配置直接定义多种音视频呼叫模型用于测试。测试系统可以通过标准的SIP协议接入不同的呼叫中心,按照模型进行测试集音视频流的发送,将接收到的流送入训练好的神经网络得到当前呼叫中心的得到视频质量的评估结果输出报告,并逐步提升发送流的路数,直到质量指标下降到设定阈值停止。
本公开实施例为各个呼叫中心系统提供音视频的处理能力的检测工具,能够提供一种基于深度学习的呼叫中心音视频处理质量和性能检测系统。为呼叫中心的音视频处理能力提供可靠的性能检测指标,帮助呼叫中心发现性能瓶颈,改进问题,提升音视频处理能力和用户体验。
本领域的普通技术人员应该知道,关于呼叫中心涉及通过深度学习及神经网络检测其音视频处理能力的系统或方案均属于本公开实施例的保护范围内。
显然,本领域的技术人员应该明白,上述的本公开实施例的各模块、单元或各步骤可以用通用的计算装置来实现,它们可以集中在单个的计算装置上,或者分布在多个计算装置所组成的网络上,它们可以用计算装置可执行的程序代码来实现,从而,可以将它们存储在存储装置中由计算装置来执行,并且在某些情况下,可以以不同于此处的顺序执行所示出或描述的步骤,或者将它们分别制作成各个集成电路模块,或者将它们中的多个模块或步骤制作成单个集成电路模块来实现。这样,本公开实施例不限制于任何特定的硬件和软件结合。
以上所述仅为本公开实施例的优选实施例而已,并不用于限制本公开实施例,对于本领域的技术人员来说,本公开实施例可以有各种更改和变化。凡在本公开实施例的原则之内,所作的任何修改、等同替换、改进等,均应包含在本公开实施例的保护范围之内。

Claims (13)

  1. 一种音视频处理性能测试方法,包括:
    构建音视频数据集,并对所述音视频数据集进行标记;
    将标记后的所述音视频数据集输入至卷积神经网络级联长短期记忆神经网络CNN+LSTM进行训练;
    配置呼叫模型的参数信息,并设置音视频数据的质量阈值和系统资源使用信息阈值;
    将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息;
    判断所述音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加所述终端的数量,并重复执行将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息的步骤,直至所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值,输出处理性能测试结果。
  2. 根据权利要求1所述的方法,其中,所述对音视频数据集进行标记,包括以下至少之一:
    将所述音视频数据集按不同劣化方式进行劣化;
    将每种所述劣化方式划分为多个等级,构建劣化评分向量;
    根据所述劣化方式和划分等级对所述音视频数据集进行标记。
  3. 根据权利要求2所述的方法,其中,所述不同劣化方式至少包括以下之一:
    视频模糊;
    视频抽帧;
    音视频不同步;
    音频加噪。
  4. 根据权利要求3所述的方法,其中,所述将每种所述劣化方式划分为多个等级,构建劣化评分向量,包括以下至少之一:
    视频模糊采用不模糊,采用3*3,5*5,9*9,15*15五种大小的卷积核进行高斯滤波;
    视频抽帧按照间隔5帧,10帧,15帧,30帧,不抽帧处理;
    音视频不同步按照-5s,-2s,0s,2s,5s进行不同步处理;
    音频加噪按照均值为零,四个递增的方差添加高斯白噪声以及不加噪处理;
    构建劣化评分向量Score[a1,a2,a3,a4],其中,a1表示视频模糊的等级分数,a2表示视频抽帧的等级分数,a3表示音视频不同步的等级分数,a4表示音频加噪的等级分数,1≤ai≤5,最高分即未劣化的视频质量得分向量为[5,5,5,5]。
  5. 根据权利要求1所述的方法,其中,所述呼叫模型的参数信息包括以下至少之一:
    呼叫路数;
    音视频数据的发送时间;
    视频分辨率;
    音视频数据的编码方式;
    音视频数据的传输码率。
  6. 根据权利要求5所述的方法,其中,在所述配置呼叫模型的参数信息之后,还包括:
    配置目标呼叫中心的接入信息,其中所述目标呼叫中心支持多个不同的所述呼叫模型;
    所述终端通过SIP协议接入所述目标呼叫中心。
  7. 一种音视频处理性能测试装置,包括:
    数据处理模块,设置为构建音视频数据集,并对所述音视频数据集进行标记;
    神经网络训练模块,设置为将标记后的所述音视频数据集输入至卷积神经网络级联长短期记忆神经网络CNN+LSTM进行训练;
    第一配置模块,设置为配置呼叫模型的参数信息,并设置音视频数据的质量阈值和系统资源使用信息阈值;
    测试模块,设置为将来自终端的音视频数据输入至所述训练好的CNN+LSTM神经网络模型,获得音视频质量信息;
    判断循环模块,设置为判断所述音视频质量信息和预先配置的系统资源使用信息是否满足小于或者等于对应的阈值,若不满足,则增加所述终端的数量,返回至所述测试模块,直至所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值;
    报告模块,设置为在所述音视频质量信息和预先配置的系统资源使用信息小于或者等于对应的阈值时,根据所述音视频质量信息和系统资源使用信息输出处理性能测试结果。
  8. 根据权利要求7所述的装置,其中,所述数据处理模块进一步包括以下至少之一:
    劣化单元,设置为将所述音视频数据集按不同劣化方式进行劣化;
    等级划分单元,设置为将每种所述劣化方式划分为多个等级,构建劣化评分向量;
    标记单元,设置为根据所述劣化方式和划分等级对所述音视频数据集进行标记。
  9. 根据权利要求8所述的装置,其中,所述不同劣化方式至少包括以下之一:
    视频模糊;
    视频抽帧;
    音视频不同步;
    音频加噪。
  10. 根据权利要求7所述的装置,其中,所述呼叫模型的参数信息包括以下至少之一:
    呼叫路数;
    音视频数据的发送时间;
    视频分辨率;
    音视频数据的编码方式;
    音视频数据的传输码率。
  11. 根据权利要求7所述的装置,其中,还包括:
    第二配置模块,设置为配置目标呼叫中心的接入信息,其中所述目标呼叫中心支持多个不同的所述呼叫模型;
    终端接入模块,设置为将所述终端通过SIP协议接入所述目标呼叫中心。
  12. 一种计算机可读存储介质,所述计算机可读存储介质中存储有计算机程序,其中,所述计算机程序被处理器执行时实现所述权利要求1至6任一项中所述的方法。
  13. 一种电子装置,包括存储器、处理器以及存储在所述存储器上并可在所述处理器上运行的计算机程序,所述处理器执行所述计算机程序时实现所述权利要求1至6任一项中所述的方法。
PCT/CN2023/094308 2022-06-06 2023-05-15 音视频处理性能测试方法及装置 WO2023236730A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210633014.1 2022-06-06
CN202210633014.1A CN117240958A (zh) 2022-06-06 2022-06-06 音视频处理性能测试方法及装置

Publications (1)

Publication Number Publication Date
WO2023236730A1 true WO2023236730A1 (zh) 2023-12-14

Family

ID=89095367

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/094308 WO2023236730A1 (zh) 2022-06-06 2023-05-15 音视频处理性能测试方法及装置

Country Status (2)

Country Link
CN (1) CN117240958A (zh)
WO (1) WO2023236730A1 (zh)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130286879A1 (en) * 2009-06-12 2013-10-31 Cygnus Broadband, Inc. Video streaming quality of experience degradation control using a video quality metric
CN104809058A (zh) * 2011-06-17 2015-07-29 阿里巴巴集团控股有限公司 一种在测试执行中动态调度并发数量的方法及装置
CN105791985A (zh) * 2014-12-22 2016-07-20 深圳Tcl数字技术有限公司 媒体流数据下载方法及装置
CN108632645A (zh) * 2017-03-17 2018-10-09 北京京东尚科信息技术有限公司 信息呈现方法和装置
CN111863033A (zh) * 2020-07-30 2020-10-30 北京达佳互联信息技术有限公司 音频质量识别模型的训练方法、装置、服务器和存储介质
CN112261437A (zh) * 2020-11-19 2021-01-22 贝壳技术有限公司 音视频质量评价方法和装置、可读存储介质、电子设备
CN113810683A (zh) * 2021-08-27 2021-12-17 南京信息工程大学 一种客观评估水下视频质量的无参考评价方法
WO2022112594A2 (en) * 2020-11-30 2022-06-02 Dolby International Ab Robust intrusive perceptual audio quality assessment based on convolutional neural networks

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20130286879A1 (en) * 2009-06-12 2013-10-31 Cygnus Broadband, Inc. Video streaming quality of experience degradation control using a video quality metric
CN104809058A (zh) * 2011-06-17 2015-07-29 阿里巴巴集团控股有限公司 一种在测试执行中动态调度并发数量的方法及装置
CN105791985A (zh) * 2014-12-22 2016-07-20 深圳Tcl数字技术有限公司 媒体流数据下载方法及装置
CN108632645A (zh) * 2017-03-17 2018-10-09 北京京东尚科信息技术有限公司 信息呈现方法和装置
CN111863033A (zh) * 2020-07-30 2020-10-30 北京达佳互联信息技术有限公司 音频质量识别模型的训练方法、装置、服务器和存储介质
CN112261437A (zh) * 2020-11-19 2021-01-22 贝壳技术有限公司 音视频质量评价方法和装置、可读存储介质、电子设备
WO2022112594A2 (en) * 2020-11-30 2022-06-02 Dolby International Ab Robust intrusive perceptual audio quality assessment based on convolutional neural networks
CN113810683A (zh) * 2021-08-27 2021-12-17 南京信息工程大学 一种客观评估水下视频质量的无参考评价方法

Also Published As

Publication number Publication date
CN117240958A (zh) 2023-12-15

Similar Documents

Publication Publication Date Title
CN111181801B (zh) 节点集群测试方法、装置、电子设备及存储介质
EP2767039B1 (en) Quality of user experience testing for video transmissions
CN111683273A (zh) 视频卡顿信息的确定方法及装置
WO2016188181A1 (zh) 一种测试方法、移动终端、服务器及计算机存储介质
CN104685870A (zh) 用于测试视频质量的方法、设备和系统
CN111629452A (zh) 一种数据传输控制方法、装置、存储介质及电子设备
CN110177300B (zh) 程序运行状态的监控方法、装置、电子设备和存储介质
CN112073128A (zh) 通信设备性能测试方法及装置
CN112584422B (zh) 5g终端性能获取方法和装置
CN111629024A (zh) 一种数据传输控制方法、装置、存储介质及电子设备
CN108696713B (zh) 码流的安全测试方法、装置及测试设备
CN107911668A (zh) 无线图传系统及方法
US9800490B2 (en) Testing by simulation using variations of real-time traffic
WO2023236730A1 (zh) 音视频处理性能测试方法及装置
US20120163212A1 (en) Apparatus and method for detecting abnormal traffic
CN109729061A (zh) 消息处理方法、装置、设备及可读存储介质
CN112351070A (zh) 文件传输方法、装置、存储介质及电子设备
CN114466224B (zh) 视频数据的编解码方法和装置、存储介质及电子设备
CN114553730B (zh) 一种应用识别方法、装置、电子设备及存储介质
CN110719526A (zh) 视频播放方法及装置
CN102956208B (zh) 一种统计终端画面帧率的方法、装置及系统
CN112422370B (zh) 语音通话质量的确定方法和装置
CN114339252A (zh) 一种数据压缩方法及装置
CN108259528A (zh) 一种资源缓存方法及装置
CN113806156A (zh) 测试方法、系统、设备以及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23818895

Country of ref document: EP

Kind code of ref document: A1