Switch to Cartesia from another voice API

Move from OpenAI, ElevenLabs, Google and more to Cartesia with one prompt for your coding agent.

Text to speech

Hume Octave

Hume's Octave TTS API shuts down on November 13: map /v0/tts to Cartesia /tts/bytes, compare Hume AI pricing, and swap the Python.

API shutdown: November 13, 2026Read the guide

OpenAI TTS

OpenAI TTS pricing per million characters, then /v1/audio/speech mapped to Cartesia field by field before the January 2027 shutdown, with Python.

Shutdown announced: January 6, 2027Read the guide

PlayHT

PlayHT shut down on December 31, 2025. Map the old /api/v2/tts/stream request, its voice and output format to Cartesia, then compare play.ht pricing.

Shut down: December 31, 2025Read the guide

Coqui TTS

An XTTS v2 alternative for teams self-hosting Coqui TTS: replace tts_to_file and the GPU that runs it with one Cartesia API call, with a working Python swap.

Company shut down: January 2024Read the guide

Amazon Polly

Move off Amazon Polly: map SynthesizeSpeech to Cartesia /tts/bytes, compare Polly pricing per million characters, and swap the boto3 code.

Read the guide

Azure Speech

Azure text to speech pricing next to Cartesia's, then the move off the REST API: SSML to JSON, output format names and a Python before and after.

Read the guide

Deepgram

Map a Deepgram Aura-2 /v1/speak request (model, encoding, sample_rate) to Cartesia, with Deepgram API pricing and a working Python swap.

Read the guide

ElevenLabs

ElevenLabs API pricing and limits next to Cartesia's, then every request field mapped one by one, with Python before and after.

Read the guide

Fish Audio

Map a Fish Audio /v1/tts request from the makers of Fish Speech (reference_id, model header) to Cartesia /tts/bytes, with pricing and code.

Read the guide

Google Cloud Text-to-Speech

Google Cloud Text-to-Speech pricing and free tiers by voice family next to Cartesia's, then text:synthesize mapped to /tts/bytes, with Python.

Read the guide

Inworld TTS

Move Inworld TTS from /tts/v1/voice to Cartesia /tts/bytes: parameter mapping, Inworld AI pricing next to Cartesia's, and a Python before and after.

Read the guide

Murf API

Swap the Murf AI speech generate endpoint for Cartesia /tts/bytes, with parameter mapping, Murf AI pricing and a Python before and after.

Read the guide

Rime

Replace the Rime AI Coda HTTP endpoint with Cartesia /tts/bytes: parameter mapping, Accept header vs output_format, and a Python swap.

Read the guide

Speechify API

Replace the Speechify API's /v1/audio/speech with Cartesia /tts/bytes for developers: parameter mapping, pricing and Python code.

Read the guide

WellSaid

WellSaid Labs pricing puts the API behind sales and keeps Starter and Pro English only. Map the /v1/tts/stream request to Cartesia, with a Python swap.

Read the guide

Speech to text

OpenAI Whisper API

A Whisper alternative before whisper-1 is removed: per minute pricing and file size limits, then /v1/audio/transcriptions mapped to one Cartesia batch call.

Model removal: February 26, 2027Read the guide

AssemblyAI

An AssemblyAI alternative: AssemblyAI pricing per minute next to Cartesia's, then the /v2/transcript submit-and-poll flow mapped to one Cartesia batch call.

Read the guide

Voice API, TTS API and speech to text API guides

Cartesia's voice API covers text to speech and speech to text with one API key. The guides above map a vendor's request onto it; these pages cover the APIs themselves.

Sonic TTS API

Text to speech over HTTP or a WebSocket, in 44 languages.

Explore the API

Ink speech to text API

Ink Whisper for recorded files, Ink-2 for live audio over a WebSocket.

Explore the API

Compare before you move

The guides assume you have decided. For listening tests and sourced numbers, start with a comparison.

All comparisons

Every Cartesia comparison, with listening tests and sourced figures.

Compare

Cartesia vs ElevenLabs

P90 latency and how much it swings, measured by Coval.

Compare

Cartesia vs Deepgram

Speech to text and text to speech, side by side.

Compare

Moving a voice agent instead?

These guides cover API calls. To move a whole agent, start with the agent comparisons.

ElevenLabs Agents alternative

Recreate an ElevenLabs agent on Cartesia Managed Agents.

Compare

Bland AI alternative

Recreate a Bland agent on Cartesia Managed Agents.

Compare

Explore Managed Agents

Configure and test an agent in Cartesia.

Explore

Frequently asked questions

How do I migrate to Cartesia from another voice API?

Create a Cartesia API key, then change the endpoint, auth header and field names that your vendor's guide maps. For text to speech, also pick a voice in the voice library. Each guide has the Python before and after.

Can I keep my voices when I switch?

No, voice IDs don't carry over between vendors. Pick a voice in the voice library, or clone one from a recording you have permission to use; an instant clone needs 10 seconds of audio.

Do these guides cover voice agents?

No, they cover text to speech and speech to text API calls. To move a whole agent, see the ElevenLabs Agents and Bland AI comparisons, which show how to recreate one by hand on Cartesia Managed Agents.

Get started today

Talk to an expert.

Connect with a member of our team and learn how Cartesia can help you build world-class voice experiences.

Contact Sales

Start building.

Access our models via API and bring a voice agent into production in minutes.

Try Cartesia