Cartesia

Generates speech from text input using state-space model (SSM) architecture optimized for real-time streaming, delivering time-to-first-audio in 40-90ms depending on model variant (Sonic-Turbo: 40ms, Sonic-3: 90ms). Streams audio chunks progressively to client as text is processed, enabling interactive voice agent applications with near-instantaneous speech output. Uses character-level pricing (1 credit per character) with support for 42 languages and dynamic voice control parameters.

Enables fine-grained control over emotional tone and prosodic characteristics of generated speech through inline text tokens and voice parameters. Supports explicit emotion markers like '[excited]' and '[sad]' embedded in input text, allowing dynamic emotional expression within a single speech generation request. Works in conjunction with voice selection and voice localization to modulate pitch, pace, and emotional coloring of output audio.

Implements credit-based pricing model where TTS generation costs 1 credit per character of input text, with additional credits for advanced features (voice cloning, localization, infilling). Credits are allocated monthly based on subscription tier (Free: 20K, Pro: 100K, Startup: 1.25M, Scale: 8M, Enterprise: custom) and do not roll over between months. This granular pricing model enables transparent cost prediction and prevents surprise bills.

Provides native integrations with popular voice agent frameworks (Pipecat, Rasa), real-time communication platforms (LiveKit, Tencent RTC, Twilio), and specialized voice agent services (Thoughtly, Vision Agents by Stream). Integrations handle authentication, streaming audio transport, and request/response marshaling, enabling developers to use Cartesia TTS/STT without building custom API clients.

Provides separate credit allocation for voice agent deployments through 'agent credits' distinct from model credits. Agent credits are prepaid amounts (Free: $1, Pro: $5, Startup: $49, Scale: $299, Enterprise: custom) that fund voice agent operations, enabling separate cost tracking and budget management for agent-based systems vs direct API usage. Mechanism for converting agent credits to API calls is not documented.

Supports two voice cloning modes: Instant Voice Cloning (IVC) requiring zero training credits, and Professional Voice Cloning (PVC) requiring 1M credits for one-time training plus 1.5 credits per character of generated speech. IVC uses speaker embedding extraction from reference audio to immediately synthesize speech in that voice without training. PVC trains a custom voice model on reference samples for higher quality and consistency, suitable for production voice agent deployments.

Generates laughter and other non-speech vocalizations (e.g., sighs, gasps) by embedding special tokens like '[laughter]' directly in input text. The synthesis engine recognizes these tokens and generates appropriate audio vocalizations that integrate seamlessly with surrounding speech, enabling natural conversational dynamics in voice agents and interactive media.

Enables regional accent and localization control for synthesized speech through voice localization parameters, allowing the same voice to be rendered with different regional accents or pronunciation patterns. Implemented as a one-time 225-credit cost per localization variant, suggesting a voice model fine-tuning or adaptation approach. Supports 42 languages with localization variants available for each.

Enables regeneration of specific portions of previously generated speech without re-synthesizing the entire utterance. Infilling works by accepting a partial text input and regenerating only the specified section, with a one-time 300-credit cost plus 1 credit per character of infill text. Useful for correcting errors, updating dynamic content, or adjusting specific phrases without full re-synthesis latency.

Automatically handles pronunciation of acronyms and initialisms by analyzing surrounding context to determine correct pronunciation (e.g., 'NASA' as word vs 'N-A-S-A' spelled out). The system infers pronunciation intent from context without requiring explicit markup, enabling natural speech synthesis for technical or specialized content containing frequent acronyms.

Provides real-time speech-to-text transcription via Ink-Whisper model using streaming audio input with dynamic chunking strategy. Audio is processed in variable-length segments optimized for transcription accuracy and latency, enabling continuous transcription of live audio streams without buffering entire utterances. Priced at $0.13 per hour of audio transcribed, supporting multiple languages and handling telephony artifacts.

Supports text-to-speech synthesis across 42 languages with consistent voice quality and emotional control across all languages. Each language can be synthesized with the same voice (if voice cloning is used) or language-specific voices, enabling multilingual voice agent deployments with consistent brand identity. Language support includes major languages (English, Spanish, French, German, Mandarin, Hindi, etc.) and regional variants.

Enforces concurrent TTS request limits based on subscription tier (Free: 2, Pro: 3, Startup: 5, Scale: 15, Enterprise: custom), preventing resource exhaustion and ensuring fair resource allocation across users. Concurrency limits are enforced at the API key level, with requests queued or rejected if limit is exceeded. This architecture enables predictable performance and cost control for multi-user deployments.