providers/CartesiaBatchTTS.ts

agentos · src/io/voice-pipeline/providers/CartesiaBatchTTS.ts · typescript · Apache-2.0
voice

* @module voice-pipeline/providers/CartesiaBatchTTS Batch text-to-speech via Cartesia's REST API (`POST /tts/bytes`). Implements {@link IBatchTTS} for one-shot narration synthesis. Contract notes (verified 2026-07-08): Bearer auth with the pinned `Cartesia-Version: 2026-03-01` he

// Source: https://github.com/framerslab/agentos | License: Apache-2.0
// Provided via justclaw.it.com; keep this notice with the code.
/**
 * @module voice-pipeline/providers/CartesiaBatchTTS
 *
 * Batch text-to-speech via Cartesia's REST API (`POST /tts/bytes`).
 * Implements {@link IBatchTTS} for one-shot narration synthesis.
 *
 * Contract notes (verified 2026-07-08): Bearer auth with the pinned
 * `Cartesia-Version: 2026-03-01` header (the only valid value today); prosody rides
 * `generation_config` (the vendor's top-level `speed` is deprecated — the
 * AgentOS `speed` knob maps into generation_config, clamped to Cartesia's
 * documented 0.6–1.5 range). `opus` output has no Cartesia container and
 * falls back to mp3, with the produced format reported on the result.
 * Cartesia has n

... (truncated -- full source via MCP)

See the full source, get the GitHub permalink, and search for more like it.

Get a free API key