Keep the ElevenLabs SDK. Point it at AllModels, then replace the model and voice. Your speech-generation call stays the same.
https://api.allmodels.io/el.cartesia/sonic-3.5 and choose a Cartesia voice ID.AllModels is an independent service and is not affiliated with, endorsed by, or sponsored by OpenAI, ElevenLabs, Fish Audio, or Cartesia.
These examples generate a normal MP3 response. Highlighted lines are the only provider-migration changes.
import { ElevenLabsClient } from "@elevenlabs/elevenlabs-js";
const client = new ElevenLabsClient({
apiKey: process.env.ELEVENLABS_API_KEY
});
const audio = await client.textToSpeech.convert("ELEVENLABS_VOICE_ID", {
text: "Hello from my application",
modelId: "eleven_flash_v2_5",
outputFormat: "mp3_44100_128"
});import { ElevenLabsClient } from "@elevenlabs/elevenlabs-js";
const client = new ElevenLabsClient({
apiKey: process.env.ALLMODELS_API_KEY,
baseUrl: "https://api.allmodels.io/el"
});
const audio = await client.textToSpeech.convert("CARTESIA_VOICE_ID", {
text: "Hello from my application",
modelId: "cartesia/sonic-3.5",
outputFormat: "mp3_44100_128"
});import os
from elevenlabs.client import ElevenLabs
client = ElevenLabs(
api_key=os.environ["ELEVENLABS_API_KEY"],
)
audio = client.text_to_speech.convert(
"ELEVENLABS_VOICE_ID",
text="Hello from my application",
model_id="eleven_flash_v2_5",
output_format="mp3_44100_128",
)import os
from elevenlabs.client import ElevenLabs
client = ElevenLabs(
api_key=os.environ["ALLMODELS_API_KEY"],
base_url="https://api.allmodels.io/el",
)
audio = client.text_to_speech.convert(
"CARTESIA_VOICE_ID",
text="Hello from my application",
model_id="cartesia/sonic-3.5",
output_format="mp3_44100_128",
)Switching providers through AllModels is intentionally simple: keep the SDK and request flow you already use, point the client at AllModels, then choose a different model and voice.
The AllModels-compatible ElevenLabs SDK surface supports text-to-speech and streaming TTS, file transcription, realtime transcription, Speech Engine conversations, and voice discovery. Read the complete ElevenLabs SDK compatibility guide for setup details, supported methods, and current limitations.
This guide switches speech generation to Sonic 3.5, Cartesia’s streaming text-to-speech model. Cartesia highlights natural pacing and emotional expression, accurate transcript following, clean multilingual audio, natural handling of alphanumeric text, and support for 42 languages.
Cartesia’s Ink-2 announcement covers the complementary speech-to-text side of its voice stack, not the TTS model used in the code above. Cartesia describes Ink-2 as a realtime voice-agent transcription model focused on accuracy, semantic turn detection, and low time-to-final-transcript.
Quick answers
No. Keep the installed SDK and its speech method. AllModels exposes a compatible base URL for it.
No. Choose a voice from the Cartesia Sonic 3.5 voice list, then replace CARTESIA_VOICE_ID with its ID.
Yes. Keep the AllModels base URL and API key, then change the model and voice to another supported provider.