WO2016131241A1 - 质检方法及装置 - Google Patents

质检方法及装置 Download PDF

Info

Publication number
WO2016131241A1
WO2016131241A1 PCT/CN2015/088907 CN2015088907W WO2016131241A1 WO 2016131241 A1 WO2016131241 A1 WO 2016131241A1 CN 2015088907 W CN2015088907 W CN 2015088907W WO 2016131241 A1 WO2016131241 A1 WO 2016131241A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
quality inspection
service
voice
voice data
Prior art date
Application number
PCT/CN2015/088907
Other languages
English (en)
French (fr)
Inventor
赖永森
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 WO2016131241A1 publication Critical patent/WO2016131241A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing

Definitions

  • the present invention relates to the field of communications, and in particular to a quality inspection method and apparatus.
  • the existing quality inspection methods are mostly the quality inspection by the quality inspection personnel to extract the operator's service recording or auxiliary system.
  • This kind of quality inspection method has the following drawbacks.
  • the spot check can not cover all the service processes of all customer service personnel. It requires a lot of manpower and material resources to complete the quality inspection.
  • the business and data sources are more difficult to change the quality inspection.
  • the quality inspection standards of human factors are different. .
  • an embodiment of the present invention provides a quality inspection method and apparatus.
  • a quality inspection method including: a quality inspection system acquires service data generated by an operator during a service process; and the quality inspection system converts the service data into a standby capable of detection Checking data; the quality inspection system performs quality inspection on the data to be inspected.
  • the service data includes at least one of the following: voice data, text data, video data, and operation log data.
  • the quality inspection system converts the service data into the to-be-checked data that can be detected, including:
  • the first text data and the second text data are used as the to-be-checked data.
  • the quality inspection system converts the service data into the to-be-checked data that can be detected, including:
  • the voice data is divided into the role information in the voice data according to a specified rule, including:
  • the voice data is divided into different character information according to different voiceprint information in the voice data.
  • the method further includes:
  • a quality inspection device which is applied to a quality inspection system, comprising: an acquisition module configured to acquire service data generated by an operator during a service process; and a conversion module configured to: The service data is converted into the quality inspection data to be detected; the quality inspection module is configured to perform quality inspection on the data to be inspected.
  • the service data acquired by the acquiring module includes at least one of the following: voice data, text data, video data, and operation log data.
  • the converting module when the service data includes: video data, includes: a splitting unit, configured to split the video data into first text data and voice data; a unit configured to convert the voice data into second text data; and a determining unit configured to use the first text data and the second text data as the to-be-checked data.
  • the converting module when the service data includes: voice data, includes: a first distinguishing unit, configured to distinguish different role information in the voice data according to a specified rule; and second distinguishing unit And being configured to distinguish the voice data according to the role information; and the second converting unit is configured to convert the differentiated voice data into the to-be-checked data.
  • the technical solution for quality inspection using the data of the quality inspection system during the service of the attendant service avoids excessive human participation, and solves the waste of artificial resources and quality inspection standards caused by manual quality inspection in the related technology.
  • the problem of non-unification and other issues has improved the efficiency of quality inspection and ensured the fairness of quality inspection.
  • FIG. 1 is a flow chart of a quality inspection method according to an embodiment of the present invention.
  • FIG. 2 is a block diagram showing the structure of a quality inspection apparatus according to an embodiment of the present invention.
  • FIG. 3 is a block diagram showing the structure of a conversion module 22 of a quality inspection apparatus according to an embodiment of the present invention
  • FIG. 4 is a block diagram showing another structure of the conversion module 22 of the quality inspection apparatus according to an embodiment of the present invention.
  • FIG. 5 is a structural block diagram of an apparatus of an intelligent quality inspection system according to a preferred embodiment 2 of the present invention.
  • FIG. 6 is a flow chart of module interaction according to a preferred embodiment 3 of the present invention.
  • FIG. 1 is a flowchart of a quality inspection method according to an embodiment of the present invention. As shown in FIG. 1, the method includes the following steps:
  • Step S102 the quality inspection system acquires service data generated by the operator during the service process
  • Step S104 the quality inspection system converts the service data into the to-be-checked data that can be detected
  • step S106 the quality inspection system performs quality inspection on the quality inspection data.
  • the technical solution for quality inspection using the data of the quality inspection system during the service of the service personnel avoids excessive human participation, and solves the waste of artificial resources and quality inspection standards caused by manual quality inspection in the related technology.
  • the issue of unification and other issues has improved the efficiency of quality inspection and ensured the fairness of quality inspection.
  • the service data includes at least one of the following: voice data, text data, video data, and operation log data, that is, the service data acquired by the quality inspection system includes, but is not limited to, one of the foregoing four types.
  • Other service data that may be included in the embodiment of the present invention may also be included in the scope of protection of the embodiments of the present invention.
  • step S104 there may be multiple implementation manners.
  • the example of the present invention mainly uses video data and operation logs as an example:
  • step S104 can be implemented by:
  • step 104 may have the following implementations:
  • the voice data is divided into different role information, that is, in the embodiment of the present invention, different characters are determined according to different voiceprint information in the voice data, and then the voice data is classified according to the role, and then the voice data is classified. , to achieve the quality inspection process more quickly.
  • step S106 After performing step S106, the following technical solution may also be executed: recording the quality inspection result, and issuing the above-mentioned quality inspection result, the quality inspection result may be used to analyze the working ability of the operator, etc., and may also be used to analyze which one The mode of operation of the operator is optimal, which in turn can promote this optimal way of working.
  • a quality inspection device is also provided, which is applied to the quality inspection system, and is used to implement the above-mentioned embodiments and preferred embodiments.
  • the descriptions have been omitted, and the modules involved in the device are described below. Be explained.
  • the term "module” may implement a combination of software and/or hardware of a predetermined function.
  • the apparatus described in the following embodiments is preferably implemented in software, hardware, or a combination of software and hardware, is also possible and contemplated.
  • 2 is a block diagram showing the structure of a quality inspection apparatus according to an embodiment of the present invention. As shown in Figure 2, the device comprises:
  • the obtaining module 20 is configured to obtain service data generated by the operator during the service process
  • the conversion module 22 is connected to the acquisition module 20 and configured to convert the service data into quality data to be detected that can be detected;
  • the quality inspection module 24 is connected to the conversion module 22 and configured to perform quality inspection on the above-mentioned quality inspection data.
  • the technical solution for quality inspection during the period of the quality inspection system dialogue service is used to avoid excessive human participation, and the waste of artificial resources and quality due to manual quality inspection is solved in the related technology.
  • the inspection standards are not uniform, and so on, which improves the efficiency of quality inspection and ensures the fairness of quality inspection.
  • the foregoing service data acquired by the obtaining module 20 includes at least one of the following: voice data, text data, video data, and operation log data.
  • the conversion module 22 when the service data includes video data, the conversion module 22 includes: a split unit 220, which is configured to The video data is split into the first text data and the voice data.
  • the first converting unit 222 is connected to the splitting unit 220 and configured to convert the voice data into the second text data.
  • the determining unit 224 is connected to the determining unit 224.
  • the first text data and the second text data are set as the to-be-checked data.
  • the conversion module 22 when the service data includes voice data, the conversion module 22 includes: a first distinguishing unit 226, And the second distinguishing unit 228 is connected to the first distinguishing unit 226, and is configured to distinguish the voice data according to the role information; the second converting unit 230 and the second distinguishing unit are configured to be configured to distinguish the different character information in the voice data according to the specified rule. 228 connection, set to be The differentiated voice data is converted into the above-mentioned to-be-checked data.
  • the preferred embodiment of the present invention provides a service-based intelligent quality inspection system and implementation method, which can independently learn, form an experience base, provide service guidance and reference for the operator, and continuously update the quality inspection standard, and handle various business service scenarios and Multi-channel sources of multiple types of data, combined with semantic analysis and operator trajectory analysis of user intent and attendant service capabilities, automatic quality inspection of the service process, to achieve the purpose of manual quality inspection, reduce the human participation and fully cover the services of all operators The process helps operators increase productivity and reduce costs.
  • the preferred embodiment of the present invention adopts the following technical solutions instead of manual quality inspection, and covers all data intelligent quality inspection, including the following modules:
  • the data scheduling storage module (equivalent to the acquisition module 20) is mainly configured by the attendant to dispatch the voice, text, video, operation log, etc. generated during the user service to the quality inspection system.
  • the data conversion module (corresponding to the above conversion module 22): mainly converts the original data into internal definition data of the quality inspection system, such as voice to text, video to voice and text, and the operation log can locate relevant data of key nodes of the operating system, and Through speech analysis, tone, service duration, mute duration, keyword split, etc.
  • Behavior Analysis Module Combining semantics with the user's purpose analysis, the operator service trajectory analysis provides raw data for quality inspection.
  • Learning module Through the analysis and analysis of a large number of data, form a library of experience, and provide reference for the operator to learn, guide the work, quality inspection standards.
  • Quality inspection evaluation module This module is mainly to integrate media information filtering, comprehensive evaluation in service capability, affinity, normative, and user-defined extended evaluation items, reference standard library and quality inspection scores defined by quality inspection administrators. Insufficient and faulty analysis of the service process, providing operators with suggestions for improvement and exemplary examples.
  • the embodiment of the invention updates the quality inspection standard in real time through self-learning, adapts to various complex scenes and data sources, simulates manual quality inspection, reduces the quality inspection cost, comprehensive quality inspection, and assists the customer service personnel to improve the service level.
  • FIG. 5 is a structural block diagram of an apparatus for an intelligent quality inspection system according to a preferred embodiment 2 of the present invention. As shown in FIG. 5, the method includes:
  • the data scheduling storage module 50 further includes: a database and a data scheduling sub-module (shown in the figure).
  • the database is mainly for storing data; the number scheduling module mainly collects and integrates the original quality inspection data, which may be real-time data or historical data, and is scheduled and converted into quality inspection data according to internal definitions, and the source thereof includes the service process. Generated but not limited to voice, text, video, operation log, etc.
  • the data conversion module 52 a video conversion engine, a voice conversion engine, and a log extraction engine.
  • the video conversion engine mainly separates images and sounds.
  • the voice conversion engine mainly splits the voice into user and operator voices by role and converts them. Into the text, at the same time generate tone, intensity, mute duration, service duration, statement split into words, to facilitate business to do keyword matching.
  • the log extraction engine separates the operation class logs from the logs or the internal service related logs. The operation logs are converted into internal data in time series to facilitate the next module analysis.
  • the module further includes a semantic analysis engine and a process data analysis module (not shown).
  • the semantic analysis engine (which can be implemented by the prior art) mainly analyzes the content of the user after the split, and analyzes the purpose of the user, and whether the operator answers clearly and satisfactorily.
  • the process data analysis module is mainly to extract the operational key points of the operator, and analyze the rationality of the use of the system tools and the proficiency of the business.
  • the module is mainly to learn through a large amount of data, find the optimal, and build a library of experience. Provide reference for the operator to learn, guide work, and quality inspection standards.
  • the learning sample is extracted: the data is extracted according to the user's purpose and high satisfaction data.
  • the single-purpose optimal service analysis is carried out, that is, according to the behavior analysis module, the service process data is extracted, and the key points are combined with the purpose analysis to obtain the operation trajectory data, that is, to achieve a goal, all the operation key points of the operator are obtained through the developed algorithm. Out of the best.
  • the operator may go through several operation points of A-->B-->C, which takes 30S, and some operators operate the key point is A-->C, total consumption.
  • A-->C A-->C
  • the system recommends several relatively good programs to achieve this goal.
  • the shortest path is the least time-consuming, and the administrator can manually intervene in these programs through the management station. Set one of them to be optimal.
  • the single-purpose optimal service plan is entered into the experience base.
  • Quality inspection evaluation module 58 This module further includes several modules for capability evaluation, affinity evaluation, normative evaluation, extended evaluation and comprehensive evaluation. Among them, the capability evaluation mainly focuses on understanding ability, problem solving ability, post processing ability, information accuracy, and business data level. Affinity evaluation focuses on courtesy (courtesy, forbidden), enthusiasm, and patience. Normative evaluation is mainly concerned with the beginning of the term, dialect, process specifications, fill-in specifications and so on. The extended evaluation is mainly to conduct quality inspections by the quality inspection administrator's customized evaluation items. The comprehensive quality inspection is mainly based on the weighting rules set by the administrator, the formula and other scoring rules combined with the experience database, and the scores of each evaluation module are aggregated and converted to obtain the final evaluation result.
  • FIG. 6 is a flow chart of module interaction according to a preferred embodiment 3 of the present invention. As shown in FIG. 6, the method includes:
  • Step S602 The original data is extracted by the data scheduling storage module 50, including but not limited to video, audio recording, text, operation log, etc., and the first simple summary is performed, and the data structure defined in the internal quality of the quality inspection is stored in the database. ;
  • Step S604 Each sub-module of the data conversion module 52 processes the scheduled data, and the video conversion engine separates the image and the sound.
  • the voice conversion engine splits the voice according to the user and the attendant, and the text is replaced with the voice tone, strength, mute duration, service duration, and sentence segmentation.
  • the log extraction engine separates the operation class, business class, and system class log, and focuses on the operation log. It re-refining and assembling according to the internally defined data structure to facilitate the next module to perform the process reduction analysis.
  • Step S606 The behavior analysis module 54 processes the data converted in the previous step. Semantic analysis engine service dialogue content analysis, service effect clear and satisfactory analysis; process data analysis module extracts the previous log extraction engine converted data, analyzes the operation key points, simulates the restoration service process, analyzes the service attendant tool Rationality and business proficiency.
  • Step S608 The learning module 56 learns through a large amount of data, finds the optimality, and builds a library of experiences. According to the user's purpose and high satisfaction, the data processed by the data conversion module and the behavior analysis module are extracted, and the single-purpose optimal service analysis is performed, and the experience database is selected in the case of automatic or manual intervention by the system.
  • Step S610 After the data conversion and processing in the above steps, the quality verification evaluation module 58 is provided with the necessary quantifiable indicator data, and the step is to perform capability evaluation, affinity evaluation, normative evaluation on the data according to certain rules. The evaluation is extended to obtain the scores of each sub-evaluation item, and then the comprehensive evaluation sub-module performs the final scoring according to the set formula and weight.
  • Step S612 The operator or the quality inspector performs the result inquiry and the subsequent business processing through the quality inspection system.
  • the embodiments of the present invention achieve the following technical effects: solving the problems of wasted labor resources and inconsistent quality inspection standards caused by manual quality inspection, thereby improving quality inspection efficiency and ensuring quality inspection. Fairness.
  • a storage medium is further provided, wherein the software includes the above-mentioned software, including but not limited to: an optical disk, a floppy disk, a hard disk, an erasable memory, and the like.
  • modules or steps of the present invention described above can be implemented by a general-purpose computing device that can be centralized on a single computing device or distributed across a network of multiple computing devices. Alternatively, they may be implemented by program code executable by the computing device such that they may be stored in the storage device by the computing device and, in some cases, may be different from the order herein.
  • the steps shown or described are performed, or they are separately fabricated into individual integrated circuit modules, or a plurality of modules or steps thereof are fabricated as a single integrated circuit module.
  • the invention is not limited to any specific combination of hardware and software.
  • the above technical solution provided by the invention can be applied to the technical solution of quality inspection during the quality inspection process by using the data of the quality inspection system during the service of the service personnel, thereby avoiding excessive human participation and solving the related art, due to the artificial quality
  • the waste of labor resources caused by inspection and the inconsistency of quality inspection standards have further improved the quality inspection quality and ensured the fairness of quality inspection.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Marketing (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Monitoring And Testing Of Exchanges (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本发明提供了一种质检方法及装置。其中，该方法包括：质检系统获取话务员在服务过程中产生的服务数据；所述质检系统将所述服务数据转换成能够进行检测的待质检数据；所述质检系统对所述待质检数据进行质检。采用本发明提供的上述技术方案，解决了相关技术中，由于人工质检导致的浪费人工资源以及质检标准不统一等问题，进而提升了质检效率，保证了质检公平性。

Description

质检方法及装置 技术领域
本发明涉及通信领域，具体而言，涉及一种质检方法及装置。
背景技术
当前存在有很多质检技术，但是由于呼叫中心生产的产品的特殊性，使得呼叫中心在产品质检问题上面临着比一般企业更高的难度。
现有的质检方式多为由质检人员抽取话务员的服务录音或者辅助系统进行质检。这种质检方式存在如下弊端，抽查不能涵盖所有客服服务人员的所有服务过程，须要投入大量人力物力完成质检，业务和数据来源多变加大质检难度，人为因素质检标准把握不一。
针对相关技术中，由于人工质检导致的浪费人工资源以及质检标准不统一等问题，尚未提出有效的解决方案。
发明内容
为了解决上述技术问题，本发明实施例提供了一种质检方法及装置。
根据本发明的一个实施例，提供了一种质检方法，包括：质检系统获取话务员在服务过程中产生的服务数据；所述质检系统将所述服务数据转换成能够进行检测的待质检数据；所述质检系统对所述待质检数据进行质检。
在本发明实施例中，所述服务数据，包括以下至少之一：语音数据、文本数据、视频数据，以及操作日志数据。
在本发明实施例中，当所述服务数据包括：视频数据时，所述质检系统将所述服务数据转换成能够进行检测的待质检数据，包括：
将所述视频数据拆分为第一文本数据和语音数据，并将该语音数据转换为第二文本数据；
将所述第一文本数据和所述第二文本数据作为所述待质检数据。
在本发明实施例中，当所述服务数据包括：语音数据时，所述质检系统将所述服务数据转换成能够进行检测的待质检数据，包括：
按照指定规则区分所述语音数据中不同的角色信息；
根据角色信息区分所述语音数据；
将区分后的语音数据转换成所述待质检数据。
在本发明实施例中，将所述语音数据按照指定规则区分所述语音数据中的角色信息，包括：
按照所述语音数据中不同的声纹信息将该语音数据区分为不同的角色信息。
在本发明实施例中，所述质检系统对所述待质检数据进行质检之后，还包括：
记录质检结果，并发布所述质检结果。
根据本发明的另一个实施例，还提供了一种质检装置，应用于质检系统，包括：获取模块，设置为获取话务员在服务过程中产生的服务数据；转换模块，设置为将所述服务数据转换成能够进行检测的待质检数据；质检模块，设置为对所述待质检数据进行质检。
在本发明实施例中，所述获取模块获取的所述服务数据，包括以下至少之一：语音数据、文本数据、视频数据，以及操作日志数据。
在本发明实施例中，当所述服务数据包括：视频数据时，所述转换模块，包括：拆分单元，设置为将所述视频数据拆分为第一文本数据和语音数据；第一转换单元，设置为将该语音数据转换为第二文本数据；确定单元，设置为将所述第一文本数据和所述第二文本数据作为所述待质检数据。
在本发明实施例中，当所述服务数据包括：语音数据时，所述转换模块，包括：第一区分单元，设置为按照指定规则区分所述语音数据中不同的角色信息；第二区分单元，设置为根据角色信息区分所述语音数据；第二转换单元，设置为将区分后的语音数据转换成所述待质检数据。
通过本发明实施例，采用质检系统对话务员服务期间的数据进行质检的技术方案，避免了过多的人工参与，解决了相关技术中，由于人工质检导致的浪费人工资源以及质检标准不统一等问题，进而提升了质检效率，保证了质检公平性。
附图说明
此处所说明的附图用来提供对本发明的进一步理解，构成本申请的一部分，本发明的示意性实施例及其说明用于解释本发明，并不构成对本发明的不当限定。在附图中：
图1为根据本发明实施例的质检方法的流程图；
图2为根据本发明实施例的质检装置的结构框图；
图3为根据本发明实施例的质检装置的转换模块22的结构框图；
图4为根据本发明实施例的质检装置的转换模块22的另一结构框图；
图5为根据本发明优选实施例二的智能质检系统的装置的结构框图；以及
图6为根据本发明优选实施例三的模块交互流程图。
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发明。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。
本发明的其它特征和优点将在随后的说明书中阐述，并且，部分地从说明书中变得显而易见，或者通过实施本发明而了解。本发明的目的和其他优点可通过在所写的说明书、权利要求书、以及附图中所特别指出的结构来实现和获得。
为了使本技术领域的人员更好地理解本发明方案，下面将结合本发明实施例中的附图，对本发明实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本发明一部分的实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明保护的范围。
在本发明实施例中，还提供了一种质检方法，图1为根据本发明实施例的质检方法的流程图，如图1所示，包括以下步骤：
步骤S102，质检系统获取话务员在服务过程中产生的服务数据；
步骤S104，上述质检系统将服务数据转换成能够进行检测的待质检数据；
步骤S106，上述质检系统对待质检数据进行质检。
通过上述各个步骤，采用质检系统对话务员服务期间的数据进行质检的技术方案，避免了过多的人工参与，解决了相关技术中，由于人工质检导致的浪费人工资源以及质检标准不统一等问题，进而提升了质检效率，保证了质检公平性。
在本发明实施例中，服务数据，包括以下至少之一：语音数据、文本数据、视频数据，以及操作日志数据，即质检系统获取的服务数据包括但不限于上述四种中的一种，在本发明实施例中可以还包括话务员可能出现的其他服务数据均在本发明实施例的保护范围中。
对于上述步骤S104的实现方式，可以有多种实现方式，本发明示例主要以视频数据和操作日志为例进行说明：
当上述服务数据包括：视频数据时，步骤S104可以通过以下方式实现：
将上述视频数据拆分为第一文本数据和语音数据，并将该语音数据转换为第二文本数据，将上述第一文本数据和上述第二文本数据作为上述待质检数据，即将视频数据先转化为语音数据和文本数据以后，将语音数据转换为文本数据，即最终实现将所有的视频数据转换为文本数据，进而将最终确定的文本数据作为待质检数据。
当上述服务数据包括：语音数据时，步骤104可以有以下实现方式：
按照指定规则区分上述语音数据中不同的角色信息；根据角色信息区分上述语音数据；将区分后的语音数据转换成上述待质检数据，其中，按照上述语音数据中不同的声纹信息将 该语音数据区分为不同的角色信息，也就是说，在本发明实施例中，是根据语音数据中不同的声纹信息来判断不同的角色，进而对语音数据按照角色分类，进而对语音数据分类，更加快速的实现质检过程。
其中，在执行步骤S106之后，还可以执行以下技术方案：记录质检结果，并发布上述质检结果，上述质检结果可以用来分析话务员的工作能力等等方面，还可以用来分析出哪个话务员的工作方式是最优的，进而可以推广这个最优的工作方式。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本发明并不受所描述的动作顺序的限制，因为依据本发明，某些步骤可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本发明所必需的。
在本实施例中还提供了一种质检装置，应用于质检系统，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，下面对该装置中涉及到的模块进行说明。如以下所使用的，术语“模块”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2为根据本发明实施例的质检装置的结构框图。如图2所示，该装置包括：
获取模块20，设置为获取话务员在服务过程中产生的服务数据；
转换模块22，与获取模块20连接，设置为将上述服务数据转换成能够进行检测的待质检数据；
质检模块24，与转换模块22连接，设置为对上述待质检数据进行质检。
通过上述各个模块的综合作用，采用质检系统对话务员服务期间的数据进行质检的技术方案，避免了过多的人工参与，解决了相关技术中，由于人工质检导致的浪费人工资源以及质检标准不统一等问题，进而提升了质检效率，保证了质检公平性。
其中，获取模块20获取的上述服务数据，包括以下至少之一：语音数据、文本数据、视频数据，以及操作日志数据。
图3为根据本发明实施例的质检装置的转换模块22的结构框图，如图3所示，当上述服务数据包括：视频数据时，转换模块22，包括：拆分单元220，设置为将上述视频数据拆分为第一文本数据和语音数据；第一转换单元222，与拆分单元220连接，设置为将该语音数据转换为第二文本数据；确定单元224，与确定单元224连接，设置为将上述第一文本数据和上述第二文本数据作为上述待质检数据。
图4为根据本发明实施例的质检装置的转换模块22的另一结构框图，如图4所示，当上述服务数据包括：语音数据时，转换模块22，包括：第一区分单元226，设置为按照指定规则区分上述语音数据中不同的角色信息；第二区分单元228，与第一区分单元226连接，设置为根据角色信息区分上述语音数据；第二转换单元230，与第二区分单元228连接，设置为将 区分后的语音数据转换成上述待质检数据。
为了更好的理解上述质检流程，以下结合优选实施例进行说明，但不用于限定本发明示例的保护范围。
优选实施例一
本发明优选实施例提供了一种基于业务的智能质检系统和实现方法，能自主学习，形成经验库，为话务员提供服务指导和借鉴，同时不断更新质检标准，处理各种业务服务场景和多渠道来源多种类型数据，结合语义分析和话务员操作轨迹分析用户意图和话务员服务能力，自动对服务过程进行质检，达到人工质检的目的，在减少人为参与同时能全面覆盖所有话务员的服务过程，帮助运营商提高工作效率和降低成本。
本发明优选实施例采用以下技术方案，代替人工质检，覆盖全部数据的智能质检，包含如下模块：
数据调度存储模块(相当于获取模块20)：主要是话务员对用户服务期间产生的语音、文字、视频，操作日志等调度到质检系统中。
数据转换模块(相当于上述转换模块22)：主要是将原始数据转换成质检系统内部定义数据，如语音转文本，视频转语音和文本，操作日志能够定位操作系统关键节点的相关数据，以及通过语音分析出语调、服务时长、静音时长，关键字拆分等。
行为分析模块：结合语义对用户的目的分析，话务员服务轨迹分析为质检提供原始数据。
学习模块：通过对大量数据采集分析，形成经验库，为话务员学习、指导工作、质检标准提供参照。
质检考评模块：本模块主要是将媒体信息整合过滤，在服务能力、亲和力、规范性，以及用户定义的扩展考评项目，参考标准库和质检管理员定义的质检计分规则进行综合评价，分析服务过程不足和过错，为话务员提供改进建议和模范示例。
本发明实施例通过自主学习，实时更新质检标准，适应各种复杂场景和数据来源，模拟人工质检，降低了质检成本，质检全面，辅助客服人员提高服务水平。
优选实施例二
图5为根据本发明优选实施例二的智能质检系统的装置的结构框图，如图5所示，包括：
数据调度存储模块50,该模块中进一步具体包括：数据库、数据调度子模块(图中为示出)。其中数据库，主要是存储数据；数调度模块，主要是对原始质检数据进行收集整合，可以是实时数据，也可是历史数据，调度并按照内部定义转换成质检的数据,其来源包含服务过程产生的但不限于语音、文字、视频，操作日志等。
数据转换模块52，视频转换引擎、语音转换引擎、日志提取引擎。视频转换引擎主要是将图像、声音的分离。语音转换引擎主要是将语音按角色拆分成用户和话务员语音，并转换 成文字，同时生成语调、强度、静音时长、服务时长、语句拆分成词，以方便业务做关键字匹配。日志提取引擎，主要在日志中分离操作类日志，还是系统内部业务相关日志,对于操作日志按照时间先后转换成内部数据，方便下个模块解析。
行为分析模块54：该模块中进一步包括语义分析引擎和过程数据分析模块(图中未示出)。语义分析引擎(可采用现有技术实现)主要是对拆分后的对话内容，分析出用户的目的，话务员是否解答清楚、满意。过程数据分析模块主要是提取话务员的操作关键点，分析其使用系统工具的合理性、业务的熟练程度。即达到一目的，熟练和不熟练的话务员在查找和使用工具是有差异的，此处就是要根据用户目的，分析出话务员服务过程数据，此处举例如操作系统交互轨迹关键点，如话务员使用多个系统的切换点，输入用户证件号码验证、通过号码查询、业务受理、在知识库查询、填写工单等等。
学习模块56：该模块最主要是通大量数据过学习，找出最优，建立经验库。为话务员学习、指导工作、质检标准提供参照。首先提取学习样本：按照用户目的、满意度高的数据进行提取。其次进行单一目的性最优服务分析，即根据行为分析模块中提取服务过程数据，对关键点结合目的分析，得出操作轨迹数据，即达到一目的，话务员所有操作关键点，通过制定的算法得出最优。此处举例但不限于：如用户咨询同一业务，话务员可能经过A-->B-->C几个操作点，总耗时30S，有的话务员操作关键点是A-->C，总耗时20S，通过大数据统计分析，系统推荐出达到这一目的几种相对优秀方案，默认情况下以最短路径最少耗时为最优,管理员可通过管理台对这几种方案人为审核干预，设定其中一种为最优。再次将单一目的最优服务方案进入经验库。
质检考评模块58：该模块中进一步包括能力评测、亲和力评测、规范性评测、扩展评测和综合评测几个模块。其中能力评测主要关注理解能力、问题解决能力、事后处理能力，信息准确性、业务数据程度。亲和力评测主要关注礼貌程度(礼貌用语、禁语)、热情程度、耐心程度。规范性评测主要是关注开始结尾语、方言、流程规范、填单规范等。扩展评测主要是更加质检管理员自定义的考评项目进行质检。综合质检主要是按照管理员设定的权重、公式等计分规则结合经验库，将各个评测模块的分值进行汇总折算，得出最终考评结果。
优选实施例三
图6为根据本发明优选实施例三的模块交互流程图，如图6所示，包括：
步骤S602：由数据调度存储模块50提取原始数据，包括但不限于视频、录音、文字、操作日志等信息，进行第一次的简单的归纳整理，转变成质检内部定义的数据结构存在数据库中；
步骤S604：数据转换模块52各个子模块分别处理调度后的数据，视频转换引擎进行图像、声音的分离。语音转换引擎按用户和话务员语音拆分，文字换成，语音的语调、强度、静音时长、服务时长、语句分词等操作。日志提取引擎分离操作类、业务类、系统类日志,重点关注操作日志，对其再按照内部定义的数据结构重新提炼组装，以方便下一模块进行操作过程还原分析。
步骤S606：行为分析模块54将上一步转换的数据进行处理。语义分析引擎服务对话内容目的分析，服务效果是否解答清楚、满意分析；过程数据分析模块提取上一步日志提取引擎转换后的数据，进行操作关键点的解析，模拟还原服务过程，分析服务中话务员工具使用合理性、业务熟练程度。
步骤S608：学习模块56通大量数据过学习，找出最优，建立经验库。按照用户目的、满意度高提取数据转换模块和行为分析模块处理后的数据，进行单一目的性最优服务分析，在系统自动或人工干预情况下入选经验库。
步骤S610：经过上面几个步骤数据转换和处理，为质检考评模块58提供了必要的可量化的指标数据，该步骤就是对这些数据按照一定规则先进行能力评测、亲和力评测、规范性评测、扩展评测，得出各个细分考评项的分值，再由综合评测子模块按照设定的公式、权重等进行最终的打分。
步骤S612：话务员或质检员通过质检系统进行结果查询及其后续业务处理。
综上所述，本发明实施例达到了以下技术效果：解决了相关技术中，由于人工质检导致的浪费人工资源以及质检标准不统一等问题，进而提升了质检效率，保证了质检公平性。
在另外一个实施例中，还提供了一种软件，该软件用于执行上述实施例及优选实施方式中描述的技术方案。
在另外一个实施例中，还提供了一种存储介质，该存储介质中存储有上述软件，该存储介质包括但不限于：光盘、软盘、硬盘、可擦写存储器等。
需要说明的是，本发明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。应该理解这样使用的对象在适当情况下可以互换，以便这里描述的本发明的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
显然，本领域的技术人员应该明白，上述的本发明的各模块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，可选地，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，本发明不限制于任何特定的硬件和软件结合。
以上所述仅为本发明的优选实施例而已，并不用于限制本发明，对于本领域的技术人员来说，本发明可以有各种更改和变化。凡在本发明的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本发明的保护范围之内。
工业实用性
本发明提供的上述技术方案，可以应用于质检过程中，采用质检系统对话务员服务期间的数据进行质检的技术方案，避免了过多的人工参与，解决了相关技术中，由于人工质检导致的浪费人工资源以及质检标准不统一等问题，进而提升了质检效率，保证了质检公平性。

Claims (10)

  1. 一种质检方法，包括：
    质检系统获取话务员在服务过程中产生的服务数据；
    所述质检系统将所述服务数据转换成能够进行检测的待质检数据；
    所述质检系统对所述待质检数据进行质检。
  2. 根据权利要求1所述的方法，其中，所述服务数据，包括以下至少之一：
    语音数据、文本数据、视频数据，以及操作日志数据。
  3. 根据权利要求2所述的方法，其中，当所述服务数据包括：视频数据时，所述质检系统将所述服务数据转换成能够进行检测的待质检数据，包括：
    将所述视频数据拆分为第一文本数据和语音数据，并将该语音数据转换为第二文本数据；
    将所述第一文本数据和所述第二文本数据作为所述待质检数据。
  4. 根据权利要求2所述的方法，其中，当所述服务数据包括：语音数据时，所述质检系统将所述服务数据转换成能够进行检测的待质检数据，包括：
    按照指定规则区分所述语音数据中不同的角色信息；
    根据角色信息区分所述语音数据；
    将区分后的语音数据转换成所述待质检数据。
  5. 根据权利要求4所述的方法，其中，将所述语音数据按照指定规则区分所述语音数据中的角色信息，包括：
    按照所述语音数据中不同的声纹信息将该语音数据区分为不同的角色信息。
  6. 根据权利要求1所述的方法，其中，所述质检系统对所述待质检数据进行质检之后，还包括：
    记录质检结果，并发布所述质检结果。
  7. 一种质检装置，应用于质检系统，包括：
    获取模块，设置为获取话务员在服务过程中产生的服务数据；
    转换模块，设置为将所述服务数据转换成能够进行检测的待质检数据；
    质检模块，设置为对所述待质检数据进行质检。
  8. 根据权利要求7所述的装置，其中，所述获取模块获取的所述服务数据，包括以下至少之一：
    语音数据、文本数据、视频数据，以及操作日志数据。
  9. 根据权利要求8所述的装置，其中，当所述服务数据包括：视频数据时，所述转换模块，包括：
    拆分单元，设置为将所述视频数据拆分为第一文本数据和语音数据；
    第一转换单元，设置为将该语音数据转换为第二文本数据；
    确定单元，设置为将所述第一文本数据和所述第二文本数据作为所述待质检数据。
  10. 根据权利要求8所述的装置，其中，当所述服务数据包括：语音数据时，所述转换模块，包括：
    第一区分单元，设置为按照指定规则区分所述语音数据中不同的角色信息；
    第二区分单元，设置为根据角色信息区分所述语音数据；
    第二转换单元，设置为将区分后的语音数据转换成所述待质检数据。
PCT/CN2015/088907 2015-07-30 2015-09-02 质检方法及装置 WO2016131241A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201510460560.X 2015-07-30
CN201510460560.XA CN106708827A (zh) 2015-07-30 2015-07-30 质检方法及装置

Publications (1)

Publication Number Publication Date
WO2016131241A1 true WO2016131241A1 (zh) 2016-08-25

Family

ID=56689246

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/088907 WO2016131241A1 (zh) 2015-07-30 2015-09-02 质检方法及装置

Country Status (2)

Country Link
CN (1) CN106708827A (zh)
WO (1) WO2016131241A1 (zh)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108876406A (zh) * 2018-06-28 2018-11-23 中国建设银行股份有限公司 客服行为分析方法、装置、服务器及可读存储介质
CN109711996A (zh) * 2018-08-17 2019-05-03 深圳壹账通智能科技有限公司 保单双录文件质检方法、装置、设备及可读存储介质
CN109682411A (zh) * 2018-08-20 2019-04-26 深圳壹账通智能科技有限公司 产品质检方法、装置、设备及计算机可读存储介质
CN109660679B (zh) * 2018-09-27 2021-09-21 深圳壹账通智能科技有限公司 催收坐席端的监控方法、装置、设备和存储介质
CN111784277B (zh) * 2020-05-22 2023-03-24 贵州电网有限责任公司 一种it客服工单质检分析方法

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102625005A (zh) * 2012-03-05 2012-08-01 广东天波信息技术股份有限公司 具有服务质量实时监督功能的呼叫中心系统及其实现方法
CN104301554A (zh) * 2013-07-18 2015-01-21 中兴通讯股份有限公司 一种对客服人员的服务质量进行质检的装置及方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102625005A (zh) * 2012-03-05 2012-08-01 广东天波信息技术股份有限公司 具有服务质量实时监督功能的呼叫中心系统及其实现方法
CN104301554A (zh) * 2013-07-18 2015-01-21 中兴通讯股份有限公司 一种对客服人员的服务质量进行质检的装置及方法

Also Published As

Publication number Publication date
CN106708827A (zh) 2017-05-24

Similar Documents

Publication Publication Date Title
CN112804400B (zh) 客服呼叫语音质检方法、装置、电子设备及存储介质
WO2016131241A1 (zh) 质检方法及装置
CN112580367B (zh) 一种话务质检方法及装置
CN110135669A (zh) 一种服务质检管理方法及系统
CN105184315A (zh) 一种质检处理方法及系统
CN106777257B (zh) 基于话术的智能对话模型的构建系统及方法
CN107886951B (zh) 一种语音检测方法、装置及设备
US11461863B2 (en) Idea assessment and landscape mapping
CN112261230B (zh) 快递电话接听率考核方法、装置、设备、系统及存储介质
CN112885376A (zh) 一种提高语音通话质检效果的方法和装置
CN111182162A (zh) 基于人工智能的电话质检方法、装置、设备和存储介质
CN108257594A (zh) 一种会议系统及其信息处理方法
CN105810205A (zh) 一种语音处理方法及装置
CN110705309A (zh) 服务质量评测方法及系统
CN112397061A (zh) 线上互动方法、装置、设备和存储介质
CN110111011B (zh) 一种教学质量监管方法、装置及电子设备
CN110287305B (zh) 一种基于自然语言处理的智能问答管理系统
CN111045902A (zh) 服务器的压力测试方法及装置
CN113505606B (zh) 一种培训信息获取方法、装置、电子设备及存储介质
CN114091477A (zh) 一种数据处理方法及装置
CN116828109A (zh) 一种电话客服服务质量智能评估方法及系统
CN114186048A (zh) 基于人工智能的问答回复方法、装置、计算机设备及介质
WO2022000141A1 (zh) 基于机器人流程自动化rpa的疫情排查方法和装置
CN110852799A (zh) 基于意图标签的用户筛选方法、装置、电子设备及介质
CN112992150B (zh) 话术模板使用效果评价方法及装置

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: 15882380

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15882380

Country of ref document: EP

Kind code of ref document: A1