About Cartesia
Cartesia builds speech models and voice agents that run fast enough for live conversation. Its text-to-speech, speech-to-text and agent products are sold mainly to developers and companies that use an API.
What it does
Cartesia offers Sonic for text-to-speech, Ink for streaming speech-to-text, and Managed Agents for phone calls. The models are built on state space architectures. The company offers regional cloud endpoints, as well as on-premise and on-device options for teams with data residency or latency needs. Teams can build voice products on the API or ask sales for an enterprise deployment.
Key features
- Streaming text-to-speech with Sonic
- Streaming speech-to-text with Ink
- Managed voice agents for phone calls
- Regional cloud endpoints
- On-premise and on-device options
Who it's for
Engineering teams building customer service bots, call automation and voice features in apps. Enterprises that need in-region processing or on-site hosting will find the deployment options relevant.
Pricing and good to know
Managed Agents are listed with a per-minute price on the site, and there is a contact sales route for larger deals. Check the pricing page for current rates. Latency and cost depend on the model, region and call volume.
Highlights
- Generate speech from text in real time
- Transcribe streaming audio to text
- Run voice agents that answer phone calls
- Deploy in a cloud region or on-premise
- Call the models through an API
