Cartesia launches sub-100 ms voice AI model on AWS
What is Building ultra-low-latency voice AI applications on AWS?
Cartesia’s Sonic model converts text to speech and speech to text in under 100 milliseconds worldwide. It can be deployed on Amazon SageMaker or on your own EC2 GPU instances using the provided examples.
Who it is for
If you need real-time voice interaction for assistants, call-center bots, or games, you can now add fast speech capabilities without building a custom stack.