Use Case Intelligence RSS Search use cases Search
NVIDIA Telecommunications United States

T-Mobile

Customer Service Automation

T-Mobile needed more accurate, lower-latency call transcripts so its Expert Assist software could recommend answers in real time without making customers repeat themselves in noisy service interactions.

Read the original story Explore all use cases
10 percent higher accuracy
10X overall latency improvement
thousands of concurrent users

Reported impact

Business outcome

T-Mobile improved transcript accuracy by 10 percent, reduced latency by 10X, and delivered real-time speech assistance for thousands of concurrent users while preserving its customer-service experience.

  • Customer Experience
  • Productivity Improvement
  • Scalability

Solution

T-Mobile embedded NVIDIA speech AI into its Expert Assist workflow so frontline agents could see reliable transcripts and recommendations while calls were still in progress. Audio from customer conversations is processed on NVIDIA A100 Tensor Core GPUs with NVIDIA Riva handling low-latency automatic speech recognition. T-Mobile engineers used NVIDIA NeMo to fine-tune speech models on proprietary customer-care data so the system could recognize product jargon, accents, and noisy call-center conditions. The live transcripts feed Expert Assist, which recommends next-best actions and helps agents resolve issues faster. The stack is designed for thousands of concurrent users, combining custom model training with GPU-accelerated inference to keep accuracy high while maintaining real-time responsiveness in production.

Data flow

Customer-call audio is processed by Riva on NVIDIA GPUs, custom NeMo-tuned ASR models generate transcripts, and Expert Assist uses those transcripts to surface recommendations to frontline agents during the conversation.

Solution architecture

3 components · 2 layers
  1. Compute
    • NVIDIA A100 Tensor Core GPU Provides the accelerated compute used to run T-Mobile's speech models for live customer-care workloads.
    • NVIDIA NeMo Lets T-Mobile fine-tune ASR models on custom customer-service data so jargon and noisy environments are handled more accurately.
  2. Serving
    • NVIDIA Riva Runs low-latency speech recognition in production so Expert Assist can transcribe calls in real time.

Architecture clues

  • ASR models were fine-tuned on custom datasets to interpret customer jargon across noisy environments
  • Live transcripts drive recommendations inside Expert Assist for frontline agents

Evidence from the source

10 percent higher accuracy
10X overall latency improvement
These NVIDIA tools enabled T-Mobile engineers to fine-tune ASR models on their custom datasets and interpret customer jargon accurately across noisy environments.
This AI-based software transcribes real-time customer conversations and recommends solutions to thousands working on the frontline.
Transparency Evidence & methodology 100/100 reference value · 89% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value100/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    2 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence89%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    4 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. NVIDIA

    T-Mobile customer story

    First-party
    https://www.nvidia.com/en-us/case-studies/speech-ai-for-award-winning-customer-care/
    Source type
    Vendor-published customer story
    Published
    Aug 29, 2025
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Aug 29, 2025 · NVIDIA · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.

Related

Similar use cases

See all like this
NVIDIA Aug 25, 2025 Singtel Telecommunications Singapore Singtel needed to make its 5G enterprise offering compelling for AI-heavy customers, but real-time video analytics, AR/VR, and chatbot workloads required low-latency edge compute, shared GPU capacity, and… Customer Service AutomationReal-time Analytics NVIDIA A40 GPUs · NVIDIA AI Enterprise · NVIDIA T4 GPUs NVIDIA Archived · Apr 25, 2025 AT&T Telecommunications United States AT&T needed to scale AI-powered customer-care agents without letting latency, operating cost, stale documentation, or model drift degrade the quality of service delivered across a massive and constantly changing support… 84% decrease in call center analytics cost Customer Service Automation NVIDIA AI Enterprise · NVIDIA NIM · NVIDIA NeMo Curator NVIDIA Archived · Sep 25, 2024 Amdocs Telecommunications Global Amdocs needed to help telecom providers automate complex customer and network workflows with domain-specific AI while lowering inference cost and latency enough for near-real-time production use. Reduced query latency by approximately 80 percent Customer Service AutomationIT Operations Automation Llama Nemotron · NVIDIA AI Enterprise · NVIDIA DGX Cloud