Enterprise-grade speech-to-text, text-to-speech, and voice agents are now native to your AWS workflow, eliminating infrastructure headaches and unlocking sub-second latency.
- Native Integration: Deepgram has launched native support for Amazon SageMaker AI, allowing developers to deploy streaming Speech-to-Text (STT), Text-to-Speech (TTS), and Voice Agents directly as real-time endpoints without custom orchestration.
- Enterprise Security & Speed: The solution offers sub-second latency for high-stakes environments like trading floors and contact centers, while keeping all data secure within the customer’s existing Amazon Virtual Private Cloud (VPC).
- Strategic Expansion: Backed by a multi-year Strategic Collaboration Agreement with AWS, this integration simplifies the adoption of generative AI voice technologies, with live demos set for AWS re:Invent 2025.
The landscape of enterprise communication is shifting rapidly toward automation, but the technical barriers to entry have often been high. Building voice-powered applications that are fast, accurate, and secure usually requires complex pipelines and significant infrastructure overhead. Today, that paradigm shifts as Deepgram, the world’s most realistic and real-time Voice AI platform, announces a groundbreaking native integration with Amazon SageMaker AI. This development delivers streaming, real-time speech-to-text (STT), text-to-speech (TTS), and the Voice Agent API directly through Amazon SageMaker AI real-time endpoints.
For developers and enterprise teams, this means the era of “workarounds” is over. There are no longer hoops to jump through to get high-quality voice AI running in the cloud. Instead, teams can build, deploy, and scale voice-powered applications entirely inside their existing AWS workflows. By removing the need for custom pipelines or complex orchestration, Deepgram is allowing AWS customers to focus on innovation rather than infrastructure maintenance.
Seamless Streaming and Sub-Second Latency
At the heart of this announcement is the ability to achieve native streaming via Amazon SageMaker endpoints. This technical capability ensures that data flows cleanly and efficiently through the SageMaker API, enabling the sub-second latency required for high-scale, real-time use cases.
Scott Stephenson, CEO and Co-Founder of Deepgram, emphasized the transformative nature of this integration: “Deepgram’s integration with Amazon SageMaker represents an important step forward for real-time voice AI. By bringing our streaming speech models directly into SageMaker, enterprises can deploy speech-to-text, text-to-speech, and voice agent capabilities with sub-second latency, all within their AWS environment.”
This speed is critical for industries where every millisecond counts. From bustling contact centers handling thousands of simultaneous calls to trading floors where split-second decisions define success, and live analytics platforms monitoring data in real-time, the reliability of this integration promises to be a game-changer.
Security and Compliance at the Forefront
One of the primary concerns for enterprises adopting Generative AI is data sovereignty and security. Deepgram has addressed this by ensuring that the new integration aligns with stringent data residency and compliance requirements. Customers can deploy Deepgram within their own Amazon Virtual Private Cloud (Amazon VPC) or use it as a managed service.
“Enterprise developers need to build voice AI applications at scale without compromising on speed, accuracy, or security,” Stephenson noted. By bringing state-of-the-art speech models directly into the AWS environment where companies already operate, Deepgram is making it dramatically easier for organizations to create voice experiences that truly transform customer engagement.
A Strengthening Partnership
This launch is not an isolated event but the result of a deepening relationship between the two tech giants. Deepgram is an AWS Generative AI Competency Partner and has signed a multi-year Strategic Collaboration Agreement (SCA) with AWS to accelerate enterprise adoption.
Ankur Mehrotra, General Manager for Amazon SageMaker at AWS, highlighted the mutual benefits of this collaboration: “Deepgram’s new Amazon SageMaker AI integration makes it simple for customers to bring real-time voice capabilities into their AWS workflows. By offering streaming speech-to-text and text-to-speech directly through Amazon SageMaker endpoints, Deepgram helps developers accelerate innovation while maintaining data security and compliance on AWS.”
The Proven Power of Deepgram
The confidence in this new solution is backed by Deepgram’s impressive track record. Having processed over 50,000 years of audio and transcribed over 1 trillion words, Deepgram understands voice data better than perhaps any other organization. Over 200,000 developers currently build with Deepgram’s voice-native foundational models due to their unmatched accuracy and low pricing. Whether it is technology ISVs building new platforms or large enterprises solving internal use cases, the demand for realistic, real-time voice AI is growing.