We spoke with Mr. Wataru Onoue, the division manager in charge of “CROSS,” a remote customer service system for local governments provided by Advanced Network Systems Co., Ltd. commonly known as "Anet", about the reasons behind their adoption of the AmiVoice SDK and the benefits it has delivered.

Issues/background

Challenges in communicating with older adults and people who are deaf or hard of hearing,
as well as difficulties understanding speech in noisy environments

"CROSS" is a system that enables local government branch offices and local offices to handle counter services remotely from bases such as the main office or call center. It contributes to resolving labor shortages, correcting regional disparities in service quality, and standardizing the quality of operations.
However, there were also issues such as the difficulty in communicating smoothly with elderly people and those who are deaf or hard of hearing, and difficulty in hearing in noisy environments. Therefore, in order to realize a service that everyone can use with peace of mind, we considered introducing a subtitle display function that visually conveys the conversation.

Key Drivers of the Adoption

High recognition accuracy and high security that can be used offline are highly rated

Aiming to further improve services for residents and streamline counter operations, we compared and tested several voice recognition services, and selected AmiVoice SDK, which excelled in Japanese recognition accuracy, real-time performance, and flexibility of implementation.
In addition to its high recognition accuracy, its high level of security, which allows it to operate offline, was the deciding factor in its adoption as a solution that can be safely implemented in operations that require high reliability, such as local government offices.

Introduction Results

Ensuring visitor privacy while providing accurate communication
and a reassuring customer service experience

With the inclusion of AmiVoice SDK, it is now possible to subtitle what the staff say in real time and display the text on the monitor as an optional feature. Furthermore, the subtitle display is limited to the staff's voice, and the content of what the visitor says is not recorded or displayed. This ensures accurate information transmission and a reassuring service at the counter while protecting the visitor's privacy.

In addition, even elderly people and those with hearing problems can accurately understand the content of conversations, and it is expected that visitor satisfaction will be improved by being able to visually confirm the content of conversations even in situations where it is difficult to hear, such as in noisy environments.

We believe that this collaboration will enable conversations during counter service to be converted into text in real time, thereby reducing the workload of staff in creating records of service responses and reducing human error.
Going forward, we will continue to support the digitalization of local governments through "CROSS" and promote the creation of services that achieve both convenience for local residents and ease of work for employees.