Intel / post

Deepgram's Voice Agent API adds NVIDIA Nemotron and on-prem deployment

Source: x.com
Date: Published May 27, 2026

Context

Deepgram's Voice API can use NVIDIA Nemotron to generate responses between speech recognition and speech synthesis. Its deployment article describes an available configuration with voice components in a customer's private cloud network and Nemotron provided through Amazon Bedrock.

The article separately labels a fully customer-operated Nemotron service as forthcoming. That distinction matters when an organization needs the entire conversation pipeline inside its own infrastructure: the documented available configuration still depends on a managed inference endpoint.

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
inference — Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.

Read the source x.com