Deepgram Partners for Enterprise Voice AI Infrastructure

Penguin Solutions and Dell are enhancing Deepgram's voice AI for critical business applications.

Deepgram has teamed up with Penguin Solutions and Dell to build a robust AI inference infrastructure. This collaboration aims to provide high-performance, low-latency voice AI for demanding enterprise sectors like healthcare and retail. It ensures strict service level agreements are met for generative AI adoption.

Sarah Kline

By Sarah Kline

March 18, 2026

4 min read

Deepgram Partners for Enterprise Voice AI Infrastructure

Key Facts

  • Deepgram partnered with Penguin Solutions and Dell Technologies for optimized AI inference infrastructure.
  • The infrastructure uses Dell PowerEdge servers and NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs.
  • The collaboration targets high-performance, low-latency voice AI for healthcare and retail sectors.
  • The system aims to meet strict service level agreements (SLAs) for enterprise generative AI.
  • Penguin Solutions provides architectural design, efficient deployment, and ongoing performance optimization.

Why You Care

Ever wonder if the AI voice on the other end of the line truly understands your complex medical query or your important retail request? What if slow AI responses could delay essential care or frustrate your customers?

Deepgram, a leader in voice AI, recently announced a significant partnership. They are working with Penguin Solutions and Dell Technologies to build a new infrastructure. This means faster, more reliable, and highly accurate voice AI is coming. This collaboration directly impacts how you interact with AI in essential services. It ensures your essential conversations are handled with precision and speed.

What Actually Happened

Deepgram has selected Penguin Solutions to deploy AI inference infrastructure, according to the announcement. This strategic collaboration aims to enhance enterprise voice AI capabilities. It focuses on delivering and low-latency experiences. These are crucial for mission-essential applications in sectors like healthcare and retail.

The new infrastructure combines Deepgram’s voice AI models with a purpose-built architectural design. It also includes efficient deployment and ongoing performance optimization. The company reports that this setup addresses challenges like stricter service level agreements (SLAs). These agreements require infrastructure that can ensure low latency and handle high concurrent usage. The system utilizes Dell PowerEdge servers and NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs.

Why This Matters to You

This partnership means your interactions with AI-powered systems will become smoother and more dependable. Imagine needing important medical information. You won’t have to repeat yourself to an AI assistant. The system will understand you quickly and accurately. This is particularly vital in high-stakes environments.

For example, think about a customer service scenario. If you’re trying to resolve a complex issue, and accurate AI responses are key. This new infrastructure aims to eliminate frustrating delays. It ensures the AI understands your needs the first time.

Key Infrastructure Components:

  • Dell PowerEdge XE7745 Servers: Provide the computational backbone.
  • NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs: Accelerate AI processing.
  • Dell PowerScale Storage: Manages and stores vast amounts of data efficiently.

“Deepgram is focused on delivering voice AI capabilities that meet the demanding performance, scalability, and reliability requirements of enterprise environments,” said Abe Pursell, vice president of partnerships and business creation at Deepgram. This commitment directly benefits you by ensuring AI services. How might more reliable voice AI change your daily interactions with automated systems?

The Surprising Finding

One might assume that AI models are the sole focus for voice AI. However, the surprising revelation here is the immense emphasis placed on the underlying infrastructure. The team revealed that even the most voice AI models are limited without support. Joe Castillo, vice president of sales at Penguin Solutions, highlighted this. He stated, “Modern AI workloads demand infrastructure that performs consistently and scales predictably under heavy loads, particularly for real-time inference applications like voice agents.”

This challenges the common assumption that AI software alone drives performance. Instead, the physical hardware and its deployment are equally essential. The research shows that achieving reliability requires a comprehensive, end-to-end architecture. It’s not just about the AI’s ‘brain’ but also its ‘nervous system.’ This integrated approach ensures complex voice AI capabilities are delivered reliably and accurately. It’s a testament to the fact that hardware still matters significantly in the age of AI.

What Happens Next

This collaboration sets the stage for a new era of enterprise voice AI. We can expect to see these enhanced capabilities roll out over the next 12-18 months. The initial focus will likely be on essential sectors. These include healthcare and retail, as mentioned in the release. For example, hospitals could implement more AI-driven patient intake systems. These systems would offer faster, more accurate voice interactions. Retailers might deploy AI agents capable of handling complex customer queries. These agents would provide personalized support with minimal latency.

Organizations looking to modernize their customer and employee experiences should pay close attention. The company reports that this approach offers highly accurate, real-time transcription and speech synthesis. What’s more, it maintains strict data governance and control. This means businesses can adopt AI with confidence. The industry implications are clear: a higher standard for AI infrastructure is being established. This will likely push other providers to follow suit. Your future interactions with AI will be faster and more secure.

Ready to start creating?

Create Voiceover

Transcribe Speech

Create Dialogues

Create Visuals

Clone a Voice