Voices
Every narration needs a voice id. The library holds more than 500 voices across 78 languages, each with a short sample; Creator and Pro workspaces can also narrate with the voices they have cloned in Scenema.
List voices
GET https://api.scenema.ai/v1/voices
| Query | Meaning |
|---|---|
type |
library (default) for the Scenema voice library, custom for the workspace’s cloned voices. |
q |
A description of the voice, e.g. warm British narrator or upbeat young female. Matches names, accents and how the voice sounds. |
language |
A language code, e.g. es or hi. |
gender |
male or female. |
page, perPage |
Pagination; perPage up to 100, 20 by default. |
curl -H "Authorization: Bearer $SCENEMA_API_KEY" \ "https://api.scenema.ai/v1/voices?q=calm%20teacher&language=es&perPage=3"The response holds items, total, page and perPage.
Get a voice
GET https://api.scenema.ai/v1/voices/{id} returns one voice from the library, or one of the workspace’s own. Any other id returns 404 VOICE_NOT_FOUND.
Voice fields
| Field | Meaning |
|---|---|
id |
The voice id to pass to POST /v1/narrations. |
name, description |
The voice’s name and a short description of how it sounds. |
gender |
male, female or null. |
language, accent |
The language code (en) and the accent or locale (en-ZA). |
ageGroup |
The voice’s apparent age group. |
tags, bestFor |
Descriptive tags, and the uses the voice suits best, best fit first. |
sampleUrl |
A short sample to listen to. |
avatarUrl |
A portrait to show with the voice, or null. |
custom |
true for a voice the workspace cloned. |