What problem does it solve?
This Skill helps production teams turn scripts and performed audio into polished speech deliverables while choosing the right Cartesia voice API, managing realtime and batch workflows, and avoiding rights, privacy, and integration mistakes.
Core Features & Use Cases
- Production TTS Planning: Select Sonic models, voices, endpoints, output formats, pronunciation controls, and generation settings for narration, ads, explainers, notifications, avatars, and voice agents.
- Realtime and Transformation Workflows: Design WebSocket continuation flows for streaming LLM speech, timestamped SSE generation, voice cloning and localization, and voice changer workflows that preserve performed timing and intonation.
- Rights, Security, and QA: Review consent and usage rights, protect API credentials, assess retention and Zero Data Retention boundaries, estimate credits and concurrency, maintain asset manifests, and validate pronunciation, timing, technical quality, and disclosure requirements.
- Use Case: Create scene-level WAV narration for a product launch, generate word-timed captions through a timestamp-capable endpoint, or repair a choppy voice agent by moving fragmented transcript generation to WebSocket continuations.
Quick Start
Use the cartesia-sonic skill to create a production plan for the supplied script, including model and endpoint selection, voice direction, pronunciation handling, output settings, rights and privacy checks, asset manifest fields, cost and concurrency estimates, and final QA.
