For a long time, the quality management department of the call center has carried out spot checks on the seat’s audio recordings mainly through manual methods. With the rapid increase in the volume of call center services and the emergence of multiple contact methods, manual quality inspection has been unable to meet the needs of modern contact centers. Therefore, the agent quality inspection needs to be performed efficiently and accurately, and it can only be realized by relying on advanced technology.                                                                                                         


  • Coverage

At present, the sampling rate of the industry is approximately 1-2%, which means that a large number of recordings are ignored, and that hidden values ​​or risks have not been effectively excavated.

  • Business value

Due to the relatively low coverage rate, it is impossible to analyze all data, or to accurately grasp customer needs and to seize business opportunities. The business value of quality inspectors cannot be established well.

  • Quality inspection delay

Manual quality inspection is generally conducted afterwards on a regular basis. This method is not able to locate the problem at the time when the incident occurs, neither to respond to the risk in a timely manner.

  • Work efficiency

Quality inspection customer service often requires repeated listening for each randomly selected call, resulting in inefficiency and limited problems found during working hours.

  • Quality inspection standards

 Manual sampling is limited to the perception of the business. Different people often have different conclusions about a same event. Even the same person will change their views on the same event at different times. This has caused difficulties in establishing a unified standard.

  • Quality inspection costs

Quality inspections are highly repetitive and have heavy tasks. With the increase in business volume, the company needs to invest a lot of manpower and material resources to meet the needs of quality inspections while ensuring the proportion of sample inspections.


Application Scenario

With the development of artificial intelligence technology, intelligent speech recognition, big data mining and other technologies can convert mass recording data into text and other unstructured data in real time or offline, achieving 100% coverage of voice files, greatly reducing the cost of labor quality inspection, through the text analysis and data mining technology, it is possible to achieve compliance inspection, risk warning, trend analysis, business opportunities mining.


System Structure


Functional Characteristics

Record Storage and Management

Processing large-scale voice call recognition into text indexes to support high-speed retrieval, listening, and analysis of voice data over a long period of time.

Business Modeling

Through the application of role setting, position setting, range setting and expression combination, the quality inspection behavior that conforms to the standard work flow can be converted into a system-recognizable rule, and the definition of the rule and the implementation of the system are peeled off to support the quality inspection. Flexible customization of customer service is also supported.

Intelligent Quality Inspection

According to the set quality inspection rules (including keywords, sentiment indicators, and mutes, etc.), a comprehensive automatic scoring is performed on the service capabilities and service attitudes of the artificial seats, so as to achieve a high coverage rate and solve the problem of low manual coverage.

Manual Inspection

The system will filter out low-rated calls for quality control personnel to listen to, so that the quality of manual inspection can achieve accurate targeting to form a new model of “quality inspection of intelligent quality auditing + quality inspectors target listening”, and then comprehensively grasp and enhance customer service and contact center artificial agent service quality.

Data Mining

Through the report engine and the customizable report designer, it provides customers with flexible and adaptable report content, and improves the correlation between report display and content query. It can achieve data mining from report to content query.