Cartesia

Model

Real-time voice models for text to speech and speech to text

Price: Free tier, then paid
Access: API key

About

Voice AI lab with Sonic text to speech, Ink speech to text and hosted voice agents, built for low-latency streaming. Plans include monthly credits, and speech costs about 1 credit per character. Ink 2 supports five languages.

What you can do with it

Stream speech from LLM output over a WebSocket with low latency
Transcribe live audio with built-in turn detection
Build and deploy voice agents with telephony and tool calls

Get started

Create an API key at play.cartesia.ai/keys
Export it as CARTESIA_API_KEY

Example

curl -X POST https://api.cartesia.ai/tts/bytes \
  -H "Authorization: Bearer $CARTESIA_API_KEY" \
  -H "Cartesia-Version: 2026-08-14" \
  -H "Content-Type: application/json" \
  -d '{"model_id":"sonic-3.6","transcript":"Hi there! Welcome to Cartesia Sonic.","voice":"f786b574-daa5-4673-aa0c-cbe3e8534c02","output_format":{"container":"wav","encoding":"pcm_s16le","sample_rate":44100}}' \
  --output sonic.wav

Details

Hosting: Hosted service
Available in: Worldwide
Official SDKs: JavaScript/TypeScript, Python
MCP server: Remote