Cartesia
A developer API for real-time speech generation, transcription, and production voice agents.
About Cartesia
Cartesia is a developer-focused voice AI platform for real-time speech applications. Its API serves Sonic text-to-speech and Ink speech-to-text models, while Line supports production voice agents with tools, guardrails, and a configurable language model. Sonic streams generated speech and provides pronunciation, accent, and voice-cloning controls for conversational systems, dubbing, narration, and AI avatars. Cartesia fits applications that need speech generation and transcription through one API. Pricing is usage-oriented: plans include monthly credits and prepaid agent minutes, with unlimited workspace seats and voice slots. Free provides 20K credits per month plus text to speech and speech to text. Pro costs $5 per month, includes 100K credits per month, and adds a commercial-use license. Teams should estimate credit and agent-minute consumption because allowances differ by plan. As an API product, Cartesia requires development and integration work rather than offering a finished consumer voice application.
Best for
Best use cases
An interactive agent streams a spoken response while transcribing the caller in real time.
A video workflow generates expressive multilingual narration or dubbed speech from text.
A prototype starts on included monthly credits and moves to Pro when commercial use is required.
Compare before choosing
Choose Cartesia when one speech API covering generation, transcription, and agents reduces integration work.
Compare monthly credit and prepaid agent allowances against expected traffic before selecting a plan.
Consider a finished no-code voice product instead when the team cannot own API integration and operations.
Features
Real-time voice agents
Add streamed speech and transcription to conversational agents that need responsive spoken interaction.
Speech generation
Generate expressive spoken audio for narration, dubbing, avatars, and interactive applications.
Speech transcription
Convert spoken input to text through Ink models for voice interfaces and multimodal applications.
Pros
- Combines text-to-speech, speech-to-text, and voice-agent building blocks within one developer platform.
- The $0 Free plan includes monthly credits, making it practical to prototype before paying.
- Streaming speech and pronunciation controls support responsive, customized voice experiences.
Cons
- Monthly credit and prepaid agent allowances require usage forecasting as an application scales.
- Commercial use is listed as a Pro benefit, so production licensing may require a paid plan.
- API-first adoption requires engineering integration and operational ownership rather than no-code setup.
Pricing Plans
Free
- 20K credits per month
- $1 prepaid agents per month
- Text to speech
- Speech to text
Pro
- 100K credits per month
- $5 prepaid agents per month
- Everything in Free
- Commercial-use license
* Pricing information is for reference only. Please refer to the latest announcement on Cartesia's official website.