Voices

Hear a voice say a line before assigning it.

Voices any voice — one of yours (GET /voices) or a catalog voice — saying text, or a character's own first line in a read (artifactId + character), optionally with a direction (how to say it, in plain words), and answers with a playable url. On a Cartesia voice (sonic-3 and newer) the direction is sent as one emotion from the voice's list and a pace nudge, as a line's is: sentAs says which ("Sent as: sad, 0.9x"). Free. The same words in the same voice and delivery are voiced once: a repeat (or a line a deck already voiced) is served from cache (cached: true). Assign the voice afterwards with POST /artifacts/{artifactId}/voice or pin it with PATCH /story-projects/{projectId}/cast-canon. Requires Idempotency-Key. Scope: story:write.

post/voices/preview

Request

  • Base URL: https://docs.sleeperhit.studio/api/v1
  • URL: https://docs.sleeperhit.studio/api/v1/voices/preview
  • Auth: HTTP bearer

Headers

Idempotency-Keystring required

Customer-generated stable identifier for safe retries of idempotent POST requests.

Request body

voiceIdstring required
provider'cartesia' | 'elevenlabs' | 'openai' | 'deepgram' | 'hume'

The voice's provider. Your own voices know theirs; otherwise the configured table-read provider.

textstring

The words to hear. Omitted: character's first line in the read artifactId.

artifactIdstring
characterstring
directionstring

How to say it, in plain words — the way a line's acting direction is written. On a Cartesia voice it is sent as one emotion from the voice's list (preview.sentAs).

Response

The preview.

Changes

No recorded changes to this operation across all 1 revision of this API.