阅读说明：本技术 一种电话渠道智能辅助系统 (Intelligent auxiliary system for telephone channel ) 是由 王建兵 吕静 于 2020-12-18 设计创作，主要内容包括：本发明属于电话智能辅助技术领域,尤其是一种电话渠道智能辅助系统,针对普通的坐席系统不具备话述配置功能问题,现提出以下方案,包括一种电话渠道智能辅助系统,包括来电用户和图片解析模块,其特征在于,所述来电用户与用于处理流程的金牌话述导航模块相连,所述金牌话述导航模块与业务成功模块和业务失败模块相连,所述业务成功模块和业务失败模块分别于成功总结模块和失败分析模块相连,所述成功总结模块和失败分析模块与用于业务优化的业务优化中心模块相连。本发明金牌话述导航结合NLP处理逻辑值具有一点的代表性,主要是提现在易用性和核心技术上,操作上大大的简化了工作人员在配置过程中所花费的时间。(The invention belongs to the technical field of intelligent telephone assistance, in particular to an intelligent telephone channel assistance system, which aims at the problem that a common seat system does not have a speech configuration function. The gold medal speaking navigation of the invention has a little representativeness by combining with NLP processing logic value, mainly improves the usability and the core technology, and greatly simplifies the time spent by the working personnel in the configuration process in operation.)

1. An intelligent auxiliary system for a telephone channel comprises an incoming call user and a picture analysis module, and is characterized in that the incoming call user is connected with a gold medal speaking navigation module used for processing a flow, the gold medal speaking navigation module is connected with a service success module and a service failure module, the service success module and the service failure module are respectively connected with a success summary module and a failure analysis module, the success summary module and the failure analysis module are connected with a service optimization center module used for service optimization, the incoming call user is also connected with a user image module, and the picture analysis module is connected with the gold medal speaking navigation module;

the picture analysis module comprises a picture and other format module, an OCR recognition capability module and a picture contained content verification module, the picture and other formats are connected with the OCR recognition module, the OCR recognition module is connected with the picture contained content verification, and the picture contained content verification is connected with the gold medal speaking navigation module;

the gold medal speaking navigation module comprises a voice module, an ASR voice recognition module, a natural voice processing NLU module, a dialogue management DM module, an FAQ intention module, a business process module, a node module and an NLP module, wherein the voice module is connected with the ASR voice recognition module, the ASR voice recognition module is connected with the natural voice processing NLU module, the natural voice processing NLU module is connected with the dialogue management DM module, the dialogue management DM module is connected with the FAQ intention module, the FAQ intention module is connected with the business process module, the business process module is connected with the node module, and the node module is connected with the NLP module.

2. The intelligent auxiliary system for telephone channel as claimed in claim 1, wherein the OCR recognition capability module obtains the information of the nodes and edges in the uploaded picture and then transmits the recognition result to the WEB interface canvas module for drawing and restoring the flow chart, and the whole process only takes fractions of a second to convert the picture into the flow chart format on the interface.

3. The intelligent auxiliary system for telephone channel as claimed in claim 2, wherein the NLP module exits and records progress if it does not complete the command, and the NLP module reads information and enters the next node if it completes the command, for the user's post-analysis.

4. The intelligent auxiliary system for telephone channel as claimed in claim 3, wherein the service optimization center module stores and generates configured services, data is stored in JSON format and transmitted to the server, and the server analyzes the information content included in the current process.

5. The intelligent auxiliary system for telephone channel according to claim 4, wherein after the service optimization center module is stored in the service end, how the configured service is analyzed and validated, each processing of the node needs to be matched with the next node after being fed back by the NLP module, the processing of different nodes needs different capabilities, the common node only needs to be processed by the NLP algorithm, and the complex node needs to complete the analysis of the node together with the multiple capabilities of the node.

6. The intelligent auxiliary system for telephone channel according to claim 5, wherein when the nodes of each service of said navigation module of gold medal are processed by NLP, each speech in incoming call is transcribed into text content by ASR speech recognition module, the text is processed into semantic by natural speech processing NLU module, the generated semantic is processed to FAQ intention module by dialogue management, the system can not understand the intention of user until the system finishes processing all nodes, and the service processing is finished.

7. The intelligent auxiliary system for telephone channel as claimed in claim 6, wherein said gold plate speech navigation module further comprises a quality inspection step for inspecting the contents of conversation in real time by inspecting the contents transcribed in real time, mainly aiming at dirty and unhealthy contents, and interrupting the current conversation service in real time if the dirty system is inspected.

8. The intelligent auxiliary system for telephone channels as claimed in claim 7, wherein the business optimization center module is capable of recording verification data of incoming call reason after the call is completed, so as to improve accuracy for analysis of incoming call reason.

9. The system as claimed in claim 8, wherein the FAQ intention module has a training function, so that the operation and maintenance personnel can fill the BOT model with corpora, and let the machine continuously learn and refine itself, thereby continuously improving the accuracy.

10. The intelligent telephone channel auxiliary system as claimed in claim 9, wherein the gold statement navigation module enables operation and maintenance personnel to optimize the current service according to the node distribution situation of service failure, and improve the service handling capacity from the optimization node details.

Technical Field

The invention relates to the technical field of intelligent telephone assistance, in particular to an intelligent telephone channel assistance system.

Background

The system is applied to application scenes related to a plurality of telephone channel services such as telephone customer service, sales channels, urging, and the like. The seat system can reduce high labor cost and improve the operation efficiency of customer service.

At present, artificial intelligence is gradually replacing some work with larger repeatability, and the great environment promotes the continuous improvement of the seat system. The existing mature seat system can already have the functions of voice transcription, content quality inspection, speech recommendation, incoming call reason analysis, historical call analysis, navigation and the like.

Through retrieval, the patent with the Chinese patent authorization number of CN109151226A discloses an intelligent voice customer service system, which comprises an outbound module, a reception module, an auxiliary analysis module and a training module, wherein the outbound module consists of an external communication system and an outbound database, the reception module consists of an answering system, the auxiliary analysis module consists of an auxiliary database and a product analysis system, the training module consists of a simulation dialogue system and a quality inspection system, the problems of high labor cost and low work efficiency of IVR self-service voice response of the traditional customer service hot line are solved, the traditional mode and the industrial ecology of the enterprise customer service are thoroughly overturned, the marketing modes of three stages of pre-sale marketing, intervention in sale and after-sale service of the enterprise are comprehensively changed, and a set of complete commercial closed loop of mass marketing is constructed by integrating multiple technical abilities of artificial intelligence, big data voice, communication resources and the like, the system helps enterprises break marketing dilemma, directly improves sales performance of the enterprises, and assists the enterprises in intelligent digital transformation.

The above patent also has the following disadvantages: the ordinary seat system does not have the speaking configuration function, and because the system has the flow configuration of the gold medal speaking navigation function, ordinary operators can directly draw a flow chart in the system to configure complex business scenes. The configuration process can arbitrarily drag the picture, cancel and recover, and can save the picture in the PNG format. The drawn flow chart can be immediately put into effect by clicking and storing, so that the operation efficiency is greatly improved, the operation cost is simplified, the gold medal statement configuration is developed based on canvas, a user does not need to consider the compatibility and performance problems, the operation is simple, and the dynamic import of txt, json and other formats and the uploading mode of picture ORC identification is supported to further improve the work efficiency.

Disclosure of Invention

Based on the technical problems in the background art, the invention provides an intelligent auxiliary system for a telephone channel.

The invention provides an intelligent auxiliary system for a telephone channel, which comprises an incoming call user and a picture analysis module, wherein the incoming call user is connected with a gold medal speaking navigation module used for processing a flow, the gold medal speaking navigation module is connected with a service success module and a service failure module, the service success module and the service failure module are respectively connected with a success summary module and a failure analysis module, the success summary module and the failure analysis module are connected with a service optimization center module used for service optimization, the incoming call user is also connected with a user image module, and the picture analysis module is connected with the gold medal speaking navigation module;

the picture analysis module comprises a picture and other format module, an OCR recognition capability module and a picture contained content verification module, the picture and other formats are connected with the OCR recognition module, the OCR recognition module is connected with the picture contained content verification, and the picture contained content verification is connected with the gold medal speaking navigation module;

the gold medal speaking navigation module comprises a voice module, an ASR voice recognition module, a natural voice processing NLU module, a dialogue management DM module, an FAQ intention module, a business process module, a node module and an NLP module, wherein the voice module is connected with the ASR voice recognition module, the ASR voice recognition module is connected with the natural voice processing NLU module, the natural voice processing NLU module is connected with the dialogue management DM module, the dialogue management DM module is connected with the FAQ intention module, the FAQ intention module is connected with the business process module, the business process module is connected with the node module, and the node module is connected with the NLP module.

Preferably, the OCR recognition capability module acquires information of nodes and sides in the uploaded picture, and then transmits the recognition result to the WEB interface canvas module for drawing and restoring the flow chart, and the picture can be converted into the flow chart format on the interface only within a few tenths of a second in the whole process.

Preferably, if the NLP module does not complete the command, the NLP module exits and records the progress, and the user performs the post-analysis, and if the NLP module completes the command, the NLP module reads the information and enters the next node.

Preferably, the service optimization center module stores and generates the configured service, the data is stored in a JSON format and transmitted to the server, and the server analyzes the information content included in the current process.

Preferably, after the service optimization center module is stored in the service end, how the configured service is analyzed and validated, each processing of the node needs to be matched with the next node after being fed back by the NLP module, capabilities required by different node processing are different, a common node only needs to be processed by the NLP algorithm, and complex nodes need multiple capabilities of the node to complete the analysis of the node together.

Preferably, when the nodes of each service of the gold medal speaking navigation module are processed by the NLP processing process, each sentence in the incoming call is transcribed into text content by the ASR speech recognition module, the text is processed into semantics by the natural speech processing NLU module, the generated semantics are processed to the FAQ intention module through the dialogue management, the system does not understand the intention of the user until the system reaches the point, and finally the next node is matched until all the nodes are processed, and the service processing is finished.

Preferably, the gold medal speaking navigation module is also provided with a quality inspection link, the conversation content is detected in real time by detecting the content transcribed in real time, the content is mainly aiming at dirty words and unhealthy content, and if the dirty word system is detected, the current conversation service can be interrupted in real time.

Preferably, the service optimization center module can record verification data of the reason of the incoming call after the call is completed, so as to improve the accuracy of the analysis of the reason of the incoming call.

Preferably, the FAQ intention module has a training function, and operation and maintenance personnel can fill corpora into the BOT model, so that the machine can continuously learn and perfect by self, and the accuracy is continuously improved.

Preferably, the gold medal speaking navigation module can enable operation and maintenance personnel to optimize the current service according to the node distribution condition of service failure, and improve the service handling capacity from the aspect of optimizing node details.

The beneficial effects of the invention are as follows:

1. the intelligent telephone channel auxiliary system has little representativeness by combining with the logic value processed by the NLP module through the gold plate language navigation, is mainly reflected in usability and core technology, greatly simplifies the time spent by workers in the configuration process in operation, only needs one picture or one txt file to upload a service flow without knowing the knowledge behind the service, and the problem of a user is more convenient and faster to process by the high-precision algorithm of the NLP.

2. The intelligent auxiliary system for the telephone channel can enable the system to continuously learn and perfect by self through the FAQ intention module, so that the accuracy is continuously improved, operation and maintenance personnel optimize the current service according to the node distribution condition of service failure, the service handling capacity is improved from the aspect of optimization node details, and the labor cost is reduced.

The parts of the system not involved are the same as or can be implemented using existing technology.

Drawings

FIG. 1 is a schematic diagram of an intelligent assistance system for telephone channel according to the present invention;

fig. 2 is a schematic view of the gold medal language navigation of the intelligent assistance system of the telephone channel according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments, and all other embodiments obtained by a person of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the protection scope of the present invention.

In the description of the present invention, it should be noted that the terms "top", "bottom", "one side", "the other side", "front", "back", "middle part", "inside", "top", "bottom", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, and are only for convenience of describing the present invention and simplifying the description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention; the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance; furthermore, unless expressly stated or limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, as they may be fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.

Referring to fig. 1-2, an intelligent auxiliary system for a telephone channel comprises an incoming call user and a picture analysis module, wherein the incoming call user is connected with a gold medal language navigation module used for processing a flow, the gold medal language navigation module is connected with a service success module and a service failure module, the service success module and the service failure module are respectively connected with a success summary module and a failure analysis module, the success summary module and the failure analysis module are connected with a service optimization center module used for service optimization, the incoming call user is also connected with a user image module, and the picture analysis module is connected with the gold medal language navigation module;

the picture analysis module comprises a picture and other format modules, an OCR recognition capability module and a picture contained content verification module, the picture and other formats are connected with the OCR recognition module, the OCR recognition module is connected with the picture contained content verification, and the picture contained content verification is connected with the gold medal speaking navigation module;

gold medal language navigation module includes speech module, ASR speech recognition module, natural speech processing NLU module, dialogue management DM module, FAQ intention module, business process module, node module and NLP module, speech module links to each other with ASR speech recognition module, ASR speech recognition module links to each other with natural speech processing NLU module, natural speech processing NLU module links to each other with dialogue management DM module, dialogue management DM module links to each other with FAQ intention module, FAQ intention module links to each other with business process module, business process module links to each other with node module, node module links to each other with NLP module.

In the invention, an OCR recognition capability module acquires the information of nodes and sides in the uploaded picture, and then transmits the recognition result to a WEB interface canvas drawing and restoring flow chart module, and the picture can be converted into the flow chart format on the interface only within a few tenths of a second in the whole process.

In the invention, if the NLP module does not complete the command, the NLP module exits and records the progress, the user analyzes in the later period, and if the NLP module completes the command, the NLP module reads the information and enters the next node.

In the invention, the service optimization center module stores and generates the configured service, the data is stored in a JSON format and transmitted to the server, and the server analyzes the information content contained in the current process.

In the invention, after the business optimization center module is stored in the service end, how the configured business is analyzed and validated, each node processing needs to be matched with the next node after being fed back by the NLP module, the processing required capacity of different nodes is different, the common node only needs to be processed by the NLP algorithm, and the complex node needs to complete the analysis of the node by multiple capacities of the node.

In the invention, when nodes of each service of the gold medal speaking navigation module pass through the NLP processing process, each sentence in an incoming call is transcribed into text content by the ASR speech recognition module, the text is processed into semantics by the natural speech processing NLU module, the generated semantics are processed to the FAQ intention module through dialogue management, the system can not understand the intention of a user until the system finishes processing, and finally the next node is matched until all the nodes are processed, and the service processing is finished.

In the invention, the gold medal speaking navigation module is also provided with a quality inspection link, the conversation content is detected in real time by detecting the content transcribed in real time, the method mainly aims at the dirty words and unhealthy content, and if the dirty words are detected, the system can interrupt the current conversation service in real time.

In the invention, the business optimization center module can record the verification data of the reason of the incoming call after the call is completed, thereby improving the accuracy rate for the analysis of the reason of the incoming call.

In the invention, the FAQ intention module has a training function, and operation and maintenance personnel can fill corpora into the BOT model, so that the machine can continuously learn and perfect by self, thereby continuously improving the accuracy.

In the invention, the gold medal speaking navigation module can ensure that operation and maintenance personnel optimize the current service according to the node distribution condition of service failure and improve the service handling capacity from the aspect of optimizing node details.

In the invention, the processing logic value of the gold medal language navigation and the NLP module has a bit of representativeness, which is mainly reflected in usability and core technology, the time spent by workers in the configuration process is greatly simplified in operation, only one picture or one txt file is needed without knowing the knowledge behind the service, the problem of processing users is more convenient and fast due to the high-precision algorithm of the NLP, in addition, the system can continuously learn and perfect by itself through the FAQ intention module, so that the accuracy is continuously improved, and the operation and maintenance personnel optimize the current service according to the node distribution condition of service failure, the service handling capacity is improved from the aspect of optimizing the node details, and the labor cost is reduced.

The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.