We've added a feedback function that utilizes AI generation to "AmiVoice® CQM Assist".

We have added an "AI comment function" to "AmiVoice CQM Assist (POWERED by PERSOL BUSINESS PROCESS DESIGN)," an automated report generation tool for contact center customer service quality evaluations, which uses a generating AI to automatically create feedback.
This feature automatically generates feedback for operators using AI based on a service quality evaluation report that scores evaluation metrics. This reduces the burden on managers such as supervisors in analyzing evaluation results and creating comments, thereby improving operational efficiency and standardizing evaluations.

"AmiVoice CQM Assist" is an automated tool for creating response quality evaluation reports, linked with the call quality evaluation function of "AmiVoice Communication Suite," the No. 1 AI voice recognition solution for contact centers in Japan*1.
AI voice recognition AmiVoice transcribes call data into text, which is then evaluated and scored based on rule-based criteria and keywords compliant with the "HDI Support Center Analyst (SCA) *3" standards proposed by HDI-Japan *2. Individual monitoring reports are then created for each operator. Compared to traditional monitoring, which tends to have a limited number of evaluations, this system enables automated monitoring of all calls.

To provide feedback based on evaluation metrics more efficiently and reliably, we have developed a new "AI Comment Function" that automatically generates feedback using a generation AI. This function automatically generates specific feedback in accordance with evaluation criteria based on rule-based scoring results.
By utilizing the "AI comment function," administrators no longer need to interpret the results of call quality evaluation reports from scratch, enabling consistent feedback with reduced variability in evaluation and guidance content across all calls. This significantly reduces the workload of administrators while strongly supporting improvements in operator response quality.

Please note that the information shared with the AI ​​generator is limited to the evaluation score and detected keywords from "AmiVoice CQM Assist," so no personal information contained in the call content is transmitted, allowing you to use the service with peace of mind.

■ Features of the "AI Comment Function"

1. Automate the evaluation process by utilizing rule-based and generative AI.
Based on the service quality evaluation report summarizing the rule-based scoring results of "AmiVoice CQM Assist," the generating AI automatically generates feedback for operators in accordance with the evaluation criteria. By combining quantitative evaluation based on rules with feedback comments generated by the AI, it is possible to automate the entire evaluation process from evaluation to feedback.
Please note that the prompts used for generating feedback are pre-configured by our company, so administrators do not need to individually set or adjust them.

2. Automated feedback generation enables improved operational efficiency and standardization of evaluations.
Based on the scoring results of the customer service quality evaluation report, the AI ​​generates data that organizes "which utterances and responses were evaluated positively" and "what aspects were lacking," and automatically generates practical feedback that leads to improvement, divided into categories such as "positive points," "challenges and areas for improvement," "specific actions to take next," and "points to focus on next time."
The generated feedback will be reflected in the comments section of "AmiVoice CQM Assist," reducing the effort required for administrators to create and transcribe comments.

3. Foster operator satisfaction through feedback on all calls.
By automating the process, consistent evaluation and feedback can be provided for all calls without resource limitations. This reduces inconsistencies in evaluations due to differences in interpretation and expression among managers, fostering operator satisfaction and maximizing motivation and training effectiveness.

4. Operators can review feedback and engage in self-learning.
By allowing operators to review feedback for each call, they can engage in autonomous reflection and self-learning based on consistent feedback aligned with the company's evaluation criteria. They can receive feedback from all daily interactions, which may not be fully grasped through meetings with managers alone, enabling them to autonomously improve their skills while understanding their strengths and areas for improvement in a timely manner. This leads to a stable improvement in the quality of service in daily operations, as well as more efficient training, faster integration of new employees into the workforce, and improved retention rates.

5. Ensure secure operation by not transmitting personal information to the generating AI.
The AI ​​generator only receives evaluation results information such as the score, evaluation items, and detected keywords from "AmiVoice CQM Assist"; the full text of the call and personal information are not transmitted. It utilizes an AI generator environment similar to the existing "AmiVoice Communication Suite," ensuring secure and reliable operation that balances security and convenience.

*1 Source: ITR "ITR Market View: Image and Voice Recognition Market 2024" Voice Recognition Market - Vendor-Specific Sales Share for Contact Center Operations (FY2024 Forecast)
*2 HDI-Japan (operated by Think Service Co., Ltd.) is a membership organization that provides rating benchmarking, international certification programs for support staff and support centers, and related training courses.
https://www.hdi-japan.com/
*3 A Support Center Analyst (SCA) is an international standard that outlines problem-solving, response procedures, and communication skills necessary for effective service delivery, as a support representative who is the first point of contact for customers.
https://www.hdi-japan.com/hdi/certification/tr_sca.asp