---
format: "aidr-story-markdown/v1"
id: "21dc542a15790d271db5d5dad98d4b2f024bb9171a70bb9facdc550ebb0a836b"
canonical_url: "https://aidr.today/21dc542a?lang=en"
title: "Building ultra-low-latency voice AI applications on AWS"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-11-03T16:40:54.000Z"
category: "Infra"
topics: ["inference"]
source_urls: ["https://aws.amazon.com/blogs/machine-learning/building-ultra-low-latency-voice-ai-applications-on-aws/"]
summary: "Cartesia, an AWS Partner, uses purpose-built State Space Models and AWS infrastructure for ultra-low-latency voice AI applications. This delivers sub-100ms text-to-speech and speech-to-text at global scale. The Sonic model can be deployed on Amazon SageMaker AI or self-managed Amazon EC2 GPU instances."
---

# Building ultra\-low\-latency voice AI applications on AWS

> [Open the canonical story](<https://aidr.today/21dc542a?lang=en>)

**Published:** 2026-11-03T16:40:54.000Z
**Category:** Infra
**Topics:** inference

## Summary

Cartesia, an AWS Partner, uses purpose\-built State Space Models and AWS infrastructure for ultra\-low\-latency voice AI applications\. This delivers sub\-100ms text\-to\-speech and speech\-to\-text at global scale\. The Sonic model can be deployed on Amazon SageMaker AI or self\-managed Amazon EC2 GPU instances\.

## Sources

- [Story source](<https://aws.amazon.com/blogs/machine-learning/building-ultra-low-latency-voice-ai-applications-on-aws/>)

