Thai 7-Eleven operator improves customer service with conversational AI

Handling 240,000 calls a day is no mean feat. It’s a massive challenge for any call centre. CP All, the sole licensed operator of the more than 11,000 7-Eleven stores in Thailand, has turned to conversational AI for help.

The result is a 60 percent reduction in call load, freeing up its call centre operators to provide better customer service.

Built on NVIDIA;s conversational AI platform, the convenience store chain’s customer service bots tackle frequently asked questions and track customer orders.

The system comprises NVIDIA DGX systems, the NVIDIA NeMo framework for natural language processing training, and the NVIDIA Triton Inference Server for AI inference and model deployment.

“Reducing hold time for customers is a key measure of our service performance. NVIDIA technologies offer us a 60 percent reduction in the call load that human agents must handle, allowing employees to efficiently tackle more unique and complex problems raised by customers,” said Kritima Klomnoi, Project Manager of CP All subsidiary Gosoft.

Using AI-driven automatic speech recognition services, CP All’s customer phone calls are transcribed in real time. When a customer service bot recognises a question based on the intelligent FAQ system, it answers using text-to-speech technologies. Otherwise, the AI quickly analyses and routes calls to the appropriate operator.

This is an amazing feat considering that standard Thai language consists of 21 consonants, 18 pure vowel sounds, three diphthongs and five tones.

“With the help of NeMo, we were able to quickly build and improve our AI language models, which are now optimised to understand and speak the unique Thai language,” said Areoll Wu, Deputy General Manager of CP All.

“The NeMo framework enabled a 97 percent accuracy in CP All’s Thai language models, more than tenfold the accuracy achieved previously,” he added

CP All plans to expand its AI services to more business domains and scale to millions of concurrent sessions on NVIDIA GPU inference architecture.

Tagged with: