阅读说明：本技术 一种基于nlp的客服对话质量检测方法以及系统 (Customer service dialogue quality detection method and system based on NLP ) 是由 冯纪强 伍亚洲 李飞鹏 徐晨 于 2020-07-28 设计创作，主要内容包括：本发明涉及一种基于NLP的客服对话质量检测方法,该方法的执行基于客服端、NLP服务器,所述客服端包括对话录音单元、对话录音存储单元及对话录音上传单元,所述NLP服务器包括对话录音分析单元及质量评估单元,该方法包括以下步骤：S1：所述对话录音单元录取通话语音并将传输至所述对话录音存储单元；S2：所述对话录音存储单元将通话语音保存；S3：所述对话录音上传单元将通话语音上传至NLP服务器,所述对话录音分析单元分析通话语音并从中获取质量指标,之后分析得出相应质量指标的指标值；S4：所述质量评估单元基于质量指标和质量评估模型评估出对话质量等级。本发明的方法和系统适用于提供人工客服服务的部门或公司,有效节省人力和时间成本,值的推广应用。(The invention relates to a customer service dialogue quality detection method based on NLP, the execution of the method is based on a customer service end and an NLP server, the customer service end comprises a dialogue recording unit, a dialogue recording storage unit and a dialogue recording uploading unit, the NLP server comprises a dialogue recording analysis unit and a quality evaluation unit, the method comprises the following steps: s1: the conversation recording unit records conversation voice and transmits the conversation voice to the conversation recording storage unit; s2: the conversation sound recording storage unit stores the conversation voice; s3: the conversation record uploading unit uploads the conversation voice to the NLP server, the conversation record analyzing unit analyzes the conversation voice and obtains a quality index from the conversation voice, and then index values of corresponding quality indexes are obtained through analysis; s4: the quality evaluation unit evaluates a dialog quality level based on the quality index and the quality evaluation model. The method and the system are suitable for departments or companies providing manual customer service, effectively save labor and time cost, and are worthy of popularization and application.)

1. A customer service dialogue quality detection method based on NLP is characterized in that: the execution of the method is based on a customer service end and an NLP server, wherein the customer service end comprises a conversation recording unit, a conversation recording storage unit and a conversation recording uploading unit, the NLP server comprises a conversation recording analysis unit and a quality evaluation unit, and the method comprises the following steps:

s1: the conversation recording unit records conversation voice and transmits the conversation voice to the conversation recording storage unit;

s2: the conversation sound recording storage unit stores the conversation voice;

s3: the conversation record uploading unit uploads the conversation voice to the NLP server, the conversation record analyzing unit analyzes the conversation voice and obtains a quality index from the conversation voice, and then index values of corresponding quality indexes are obtained through analysis;

s4: the quality evaluation unit evaluates a dialog quality level based on the quality index and the quality evaluation model.

2. The customer service dialog quality detection method of claim 1, characterized by:

in S2, the quality index includes whether the problem is solved or not, and customer satisfaction.

3. The customer service dialog quality detection method according to claim 1 or 2, characterized by:

at S3, the dialog quality rating is in the form of a percentile score.

4. A customer service dialogue quality detection system based on NLP is characterized in that: the system comprises a customer service end and an NLP server, wherein the customer service end comprises a conversation recording unit, a conversation recording storage unit and a conversation recording uploading unit, and the NLP server comprises a conversation recording analysis unit and a quality evaluation unit;

the conversation recording unit is used for recording conversation voice and transmitting the conversation voice to the conversation recording storage unit;

the conversation sound recording storage unit is used for storing conversation voice;

the conversation recording uploading unit is used for uploading the conversation voice to the NLP server;

the conversation sound recording analysis unit is used for analyzing the conversation voice and obtaining a quality index from the conversation voice, and then is used for analyzing to obtain an index value of the corresponding quality index;

the quality evaluation unit is used for evaluating the dialogue quality grade based on the quality index and the quality evaluation model.

5. The customer service dialog quality detection system of claim 4, wherein: the quality indexes comprise problem solving and customer satisfaction.

6. The customer service dialog quality detection system of claim 4 or 5, characterized in that: the dialog quality rating is in the form of a percentile score.

Technical Field

The invention relates to a customer service dialogue quality detection method and system based on NLP.

Background

The enterprise customer service is one of important channels for information feedback of customers, and effective customer service can improve the communication efficiency and quality, reduce the customer service cost and achieve a win-win situation for the enterprise and the customers. The telephone voice not only contains character symbol information, but also contains rich information such as emotion and emotion of a speaker. In order to control the call quality of the customer service representative, the call voice of the customer service can be analyzed, and the emotional factors are important and are important factors for reflecting the call quality. The method for detecting the call quality of the customer service center of the modern enterprise is manual quality inspection. With the increasing popularity of customer service centers, manual quality inspection shows the defects of the customer service centers, the number of people is large, the cost is only too high for detecting all calls of the customer service, and therefore the customer service can only be checked in a sampling mode, but some rules in voice are difficult to find through the sampling mode, and the decision and improvement of a management layer are not facilitated.

In the past, due to the fact that a voice recognition technology is insufficient, analysis of customer service voice mainly depends on manual monitoring, and the problems of high cost and insufficient coverage range exist.

Disclosure of Invention

In order to solve the technical problem of the above technology, the present invention provides a customer service dialogue quality detection method based on NLP, the execution of the method is based on a customer service end and an NLP server, the customer service end includes a dialogue recording unit, a dialogue recording storage unit and a dialogue recording uploading unit, the NLP server includes a dialogue recording analysis unit and a quality evaluation unit, the method includes the following steps:

s1: the conversation recording unit records conversation voice and transmits the conversation voice to the conversation recording storage unit;

s2: the conversation sound recording storage unit stores the conversation voice;

s3: the conversation record uploading unit uploads the conversation voice to the NLP server, the conversation record analyzing unit analyzes the conversation voice and obtains a quality index from the conversation voice, and then index values of corresponding quality indexes are obtained through analysis;

s4: the quality evaluation unit evaluates a dialog quality level based on the quality index and the quality evaluation model.

Further, in S2, the quality index includes whether the problem is solved, and customer satisfaction.

Further, in S3, the dialog quality ranking is in the form of a fractional number of percent.

The invention also provides a customer service dialogue quality detection system based on NLP, which comprises a customer service end and an NLP server, wherein the customer service end comprises a dialogue recording unit, a dialogue recording storage unit and a dialogue recording uploading unit, and the NLP server comprises a dialogue recording analysis unit and a quality evaluation unit;

the conversation recording unit is used for recording conversation voice and transmitting the conversation voice to the conversation recording storage unit;

the conversation sound recording storage unit is used for storing conversation voice;

the conversation recording uploading unit is used for uploading the conversation voice to the NLP server;

the conversation sound recording analysis unit is used for analyzing the conversation voice and obtaining a quality index from the conversation voice, and then is used for analyzing to obtain an index value of the corresponding quality index;

the quality evaluation unit is used for evaluating the dialogue quality grade based on the quality index and the quality evaluation model.

Further, the quality index includes whether to solve the problem and customer satisfaction.

Further, the dialog quality rating is in the form of a percentile score.

The invention has the beneficial effects that:

the method and the system are suitable for departments or companies providing manual customer service, and the customer service department can evaluate the quality of each communication session of each customer service worker by the method and the system, so that heavy manual evaluation work is omitted, labor and time cost are effectively saved, and popularization and application are facilitated.

Drawings

FIG. 1 is a schematic diagram of a customer service end architecture of the present invention;

FIG. 2 is a flow chart of the application of the method of the present invention in one embodiment.

Detailed Description

The technical solutions in the embodiments of the present invention are clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The invention provides a customer service dialogue quality detection method based on NLP, the execution of the method is based on a customer service end and an NLP server, as shown in figure 1, the customer service end comprises a dialogue recording unit, a dialogue recording storage unit and a dialogue recording uploading unit, the NLP server comprises a dialogue recording analysis unit and a quality evaluation unit, the method comprises the following steps:

s1: the conversation recording unit records conversation voice and transmits the conversation voice to the conversation recording storage unit;

s2: the conversation sound recording storage unit stores the conversation voice;

s3: the conversation record uploading unit uploads the conversation voice to the NLP server, the conversation record analyzing unit analyzes the conversation voice and obtains a quality index from the conversation voice, and then index values of corresponding quality indexes are obtained through analysis;

s4: the quality evaluation unit evaluates a dialog quality level based on the quality index and the quality evaluation model.

Further, in S2, the quality index includes whether the problem is solved, and customer satisfaction.

Further, in S3, the dialog quality ranking is in the form of a fractional number of percent.

Referring to fig. 2, in an embodiment, a customer service performs a session related to the customer service with a customer through a customer service terminal, then records the session and stores the record in a voice file, then the voice file is automatically uploaded to an NLP server, the NLP server performs quality evaluation on the record to obtain a customer service quality index and a grade, and feeds the customer service quality index and the grade back to the customer or a manager, and the customer or the manager can improve the session quality according to the content of the feedback.

The invention also provides a customer service dialogue quality detection system based on NLP, which comprises a customer service end and an NLP server, wherein the customer service end comprises a dialogue recording unit, a dialogue recording storage unit and a dialogue recording uploading unit, and the NLP server comprises a dialogue recording analysis unit and a quality evaluation unit;

the conversation recording unit is used for recording conversation voice and transmitting the conversation voice to the conversation recording storage unit;

the conversation sound recording storage unit is used for storing conversation voice;

the conversation recording uploading unit is used for uploading the conversation voice to the NLP server;

the conversation sound recording analysis unit is used for analyzing the conversation voice and obtaining a quality index from the conversation voice, and then is used for analyzing to obtain an index value of the corresponding quality index;

the quality evaluation unit is used for evaluating the dialogue quality grade based on the quality index and the quality evaluation model.

Further, the quality index includes whether to solve the problem and customer satisfaction.

Further, the dialog quality rating is in the form of a percentile score.

The method and the system are suitable for departments or companies providing manual customer service, and the customer service department can evaluate the quality of each communication session of each customer service worker by the method and the system, so that heavy manual evaluation work is omitted, labor and time cost are effectively saved, and popularization and application are facilitated.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.