NTT X Perso Interactive — Travel Experiences Transformed by the Introduction of Real-time AI Interpreters

About

Japan’s largest telecommunications carrier, NTT, along with major taxi operator, recognized a critical need to enhance communication quality to better serve the growing number of international tourists. To address the language barrier between Japanese drivers and foreign passengers, Perso Interactive introduced a tablet-based, real-time Multilingual AI interpreter and AI Human–powered travel guide solution.

This solution features a Real-time AI Interpreter that facilitates natural dialogue and provides location-based insights, effectively expanding a simple taxi ride into a rich travel experience. Through on-site POC, the solution successfully demonstrated improved communication and user experience—paving the way for future commercialization and service expansion.

Background

With the surge in global visitors, Japanese tourist taxis saw an opportunity to evolve the "ride" from mere transportation into a valuable "experience". The key was to provide natural dialogue and localized insights during the short duration of the trip to boost overall travel satisfaction.

However, communication between Japanese drivers and foreign passengers was often limited due to language barriers, making it difficult to exchange information or answer questions during the ride.

To answer the question, “Can we make a 30-minute ride the most meaningful part of a trip?”, we deployed a tablet-based AI solution that combines real-time translation with an AI Human tourism guide.

Solution

By integrating the Perso Interactive SDK/API, we developed a specialized tablet application that operates flawlessly within the vehicle environment, creating a bidirectional communication system for both drivers and passengers.

The core feature is a one-touch Multilingual interaction system, designed for ease of use even while driving. With a simple button press, the driver’s speech is instantly translated into the passenger’s language, while an AI Human avatar delivers voice guidance, ensuring immediate understanding.

To enable more natural conversations, the system leverages LLM-based multi-turn dialogue, allowing interactions to continue smoothly even in unstable network conditions. In addition, the solution integrates GPS-based location intelligence, providing real-time recommendations for nearby attractions, restaurants, and hot springs—tailored to the passenger’s current location.

Results

Since the implementation of the service, the atmosphere inside the taxis has been completely transformed.

Despite language differences, drivers and passengers were able to communicate naturally, turning travel time into a meaningful part of the journey. International passengers could freely express their thoughts and questions through real-time translation, while gaining authentic local insights directly from drivers.

Drivers can now focus on the road without language-related stress while confidently fulfilling their roles as local guides. Consequently, the taxi provides a high-quality experience comparable to having a professional local guide on board.

Transportation is no longer just a means of getting from point A to B—it has become an integral part of the travel experience, redefining the standard of mobility services.

Recommended For

This use case is ideal for:

• Tourism taxi and mobility service operators

• Transportation services requiring Multilingual communication

• On-site service managers looking to reduce communication barriers with international customers

This solution is also highly applicable and scalable for related industries, including hotel concierges, airport shuttles, and AI Kiosk deployments at major tourist hubs.

Related Article: ESTsoft Signs MOU with NTT and Nippon Thumb for Introduction of AI Human Services in Japanese Taxis… AI Humans to Be Carried in Japanese Taxis!

Frequently Asked Questions

Q. How fast is the real-time interpretation?

A. Translations are delivered almost instantly upon speech, ensuring that the conversation flow remains uninterrupted. It enables smooth and natural communication, similar to face-to-face conversations.

Q. Can it be used safely while driving?

A. Yes. For drivers, the system is designed with a one-touch button interface, allowing minimal interaction so they can stay focused on driving while still providing guidance.

Q. Can it still be used in unstable network environments?

A. Yes. The system is designed with mobility in mind, ensuring stable performance and maintaining conversation flow even in environments with unstable network conditions.