Speech AI for Award-Winning Customer Care


Expert Assist needed more accurate transcriptions to offer effective recommendations to customer care agents.



Use Case

Speech AI for Customer Service



T-Mobile is a is a mobile communications provider that delivers advanced 4G LTE and transformative 5G networks. By enabling customer care agents to solve queries quickly with highly accurate NVIDIA Riva ASR-generated transcripts, T-Mobile provides the best possible service experience to their customers. T-Mobile was awarded the top spot for customer service for several consecutive years.

Assisting Agents in Real Time

In millions of daily customer care calls worldwide, consumers expect agents to fix their issues quickly and efficiently. Enterprise leaders are looking into automated and scalable solutions to assist customer care agents in providing the best experience.

T-Mobile offers award-winning customer service in their Customer Experience Centers and aims never to have the customer repeat themself. To empower their frontline agents, T-Mobile implemented Expert Assist. This AI-based software transcribes real-time customer conversations and recommends solutions to thousands working on the frontline.

NVIDIA Solution

T-Mobile uses NVIDIA Riva, the GPU-accelerated SDK for building customized speech applications, and NVIDIA NeMo, the open-source framework for building, training, and fine-tuning state-of-the-art conversational AI models. These NVIDIA tools enabled T-Mobile engineers to fine-tune ASR models on their custom datasets and interpret customer jargon accurately across noisy environments.

T-Mobile Results


  • 10 percent higher accuracy

  • 10X overall latency improvement

The T-Mobile team fine-tuned speech AI models using NVIDIA NeMo, slashing word error rate (WER) by 10 percent across noisy production environments. They reduced latency by 10X using NVIDIA Riva, enabling the highest level of real-time performance for thousands of concurrent users.

“Using DGX Cloud and Base Command Platform’s dataset management and orchestration capabilities, our data scientists have reported a 2X speedup in running experiments.”

Neda Hantehzadeh
PhD, Director of Data Science, CCCIS