Cartesia TTS

CartesiaTTS is a Patter TTSProvider backed by Cartesia’s /tts/bytes HTTP endpoint. It streams raw PCM_S16LE chunks that drop directly into Patter’s pipeline with no transcoding. Uses the platform fetch — no vendor SDK required, works on Node 18+. The default model is sonic-3 (current GA snapshot, ~90 ms TTFB target).

This page covers Cartesia TTS. For Cartesia’s ink-whisper STT see the Cartesia STT page.

Install

npm install getpatter

Usage

Use the namespaced import (getpatter/tts/cartesia) or the flat re-export (CartesiaTTS). Both auto-resolve CARTESIA_API_KEY from the environment when apiKey is omitted.

// Namespaced import
import * as cartesia from "getpatter/tts/cartesia";

const tts = new cartesia.TTS();                                  // reads CARTESIA_API_KEY
const tts2 = new cartesia.TTS({ voice: "f786b574-daa5-4673-aa0c-cbe3e8534c02" });

// Flat alias (equivalent)
import { CartesiaTTS } from "getpatter";

const tts3 = new CartesiaTTS({ voice: "f786b574-daa5-4673-aa0c-cbe3e8534c02" });

# See python-sdk/providers/cartesia-tts for the Python twin.

In an agent:

// npx tsx example.ts
import { Patter, Twilio, DeepgramSTT, CartesiaTTS } from "getpatter";

const phone = new Patter({ carrier: new Twilio(), phoneNumber: "+15550001234" });

const agent = phone.agent({
  stt: new DeepgramSTT(),                             // DEEPGRAM_API_KEY from env
  tts: new CartesiaTTS({ voice: "f786b574-daa5-4673-aa0c-cbe3e8534c02" }),
  systemPrompt: "You are a helpful assistant.",
});

await phone.serve({ agent });

Models and rates

Cartesia TTS bills per 1,000 characters synthesized. Per-model rates (defaults from getpatter/pricing):

Model	Rate / 1k chars	Notes
`sonic-3` (default)	$0.030	GA, ~90 ms TTFB. Drop-in compatible with sonic-2 voice IDs.
`sonic-2`	$0.030	Previous flagship.
`sonic-1` / `sonic-english`	$0.030	Legacy English.
`sonic-multilingual`	$0.030	Multilingual variant.

Override via new Patter({ pricing: { cartesia_tts: { models: { "sonic-3": { price: ... } } } } }).

Languages

language: "en" by default. Sonic-3 supports 30+ languages — pass any Cartesia-supported BCP-47 code ("es", "fr", "de", "it", "pt", "ja", "zh", …). Voice cloning works across all of them — the same voice ID will speak whichever language you target.

Telephony optimization

Use the carrier-aware factories so audio reaches Patter’s telephony adapter at the carrier’s native sample rate, skipping a resample step:

import { CartesiaTTS } from "getpatter";

// Twilio: PCM @ 8 kHz directly (skips the 16k→8k resample before mulaw transcode).
const twilioTts = CartesiaTTS.forTwilio();

// Telnyx: PCM @ 16 kHz (matches Telnyx's L16/16000 default — zero transcoding).
const telnyxTts = CartesiaTTS.forTelnyx();

Options

Option	Default	Notes
`apiKey`	—	Reads from `CARTESIA_API_KEY` when omitted.
`model`	`"sonic-3"`	Any Cartesia TTS model id.
`voice`	`"f786b574-..."` (Katie)	Cartesia voice id.
`language`	`"en"`	BCP-47 code.
`sampleRate`	`16000`	8000, 16000, 22050, 24000, 44100 Hz.
`speed`	—	`"fastest" ... "slowest"` or float in `[0.6, 2.0]`.
`emotion`	—	Cartesia emotion preset.
`volume`	—	Float in `[0.5, 2.0]` (sonic-3 only).
`baseUrl`	Cartesia API	Override for proxying.
`apiVersion`	`"2025-04-16"`	Cartesia API version pin.

​Cartesia TTS

​Install

​Usage

​Models and rates

​Languages

​Telephony optimization

​Options

Cartesia TTS

Install

Usage

Models and rates

Languages

Telephony optimization

Options