ElevenLabs v4
Text-to-speech on ElevenLabs Eleven v4, with 21 preset voices and 415 ElevenLabs Voice Library voices. It is best at clean, expressive voice-over for ad scripts. The model id stays `elevenlabs-v3`.
audio · model elevenlabs-v3 · catalog model-catalog@2026-10-06.v5 · pricing customer-pricing@2026-11-05.v15.Endpoint
POST https://api.roasos.com/v1/audio/elevenlabs-v3Send your API key as a bearer token. The key needs the scope api:write. Every paid request also needs an Idempotency-Key header.
The envelope fields quote_only, quote_id, webhook_url and wait_timeout_seconds behave the same on each model endpoint. Refer to Jobs and Quotes and credits.
Good at
Voice-over for a UGC script, a product explainer, or a hook line. The 21 presets cover calm, warm, deep, British, Scottish, narrator, and documentary reads. voice takes a preset name, such as Rachel, or a voice id from the roasOS snapshot of the ElevenLabs Voice Library (415 voices, taken on 2026-10-04). Another id answers 400. Set language_code to force a language when the text is not English. similarity_boost sets how closely the take follows the voice. timestamps: true answers 400: the result does not store word timing yet. It does not clone a voice, and voice_id is not a field: send a library id in voice. It does not accept media of any kind.
Pricing
The unit is one characters. Pricing key: elevenlabs-v3.
| Request | Credits |
|---|---|
1000_characters | 1 |
2000_characters | 2 |
5000_characters | 5 |
The credits come from the roasOS pricing table. No other field changes the price. To get the exact price of your request, send it with quote_only: true first.
Modes
text-to-speech
Provider endpoints: fal-ai/elevenlabs/tts/eleven-v4.
Request fields
| Field | Type | Required | Default | Rule |
|---|---|---|---|---|
apply_text_normalization | enum | no | auto | auto, on, off |
language_code | string | no | — | matches ^[a-z]{2}$ |
quote_id | string | no | — | UUID |
quote_only | boolean | no | false | — |
similarity_boost | number | no | 0.75 | 0–1 |
stability | number | no | 0.5 | 0–1 |
text | string | yes | — | ≥ 1 character; ≤ 5,000 characters |
timestamps | boolean | no | false | — |
voice | string | no | Rachel | — |
wait_timeout_seconds | integer | no | 0 | 0–60 |
webhook_url | string | no | — | https URL; ≤ 4,096 characters |
The schema is strict. A field that is not in this table gets 400 unsupported_option. The field names are the roasOS role names. One role has one name in every family.
Media
This endpoint accepts no media.
Every other role is refused: avatar_id, elements, product_id, product_images, references, reference_images, image_url, end_image_url, scene_id, scene_image, video_urls, video_url, audio_urls, audio_url, voice_id. A refused role gets 400 unsupported_option.
https://cdn.roasos.com that your workspace owns. Any other URL gets 400 reference_media_not_workspace_owned. Refer to Reference media and uploads to learn how to put a file there.Prompt
text is the script. Limit 5,000 characters. roasOS adds no wrapper.
- Write the words as they must be spoken. Spell out abbreviations that the voice must read as words.
- Use punctuation for pacing. A comma is a short beat; a full stop is a longer beat; an ellipsis is a pause.
- Keep one sentence per idea. Long compound sentences make the read flat.
- Lower
stability(0.3) for a more expressive read; raise it (0.7) for a steady announcer read. - Keep
similarity_boostat its default (0.75). Raise it to stay closer to the voice; a high value can sound less natural. - Set
apply_text_normalization: "on"when the text has prices, dates, or phone numbers. - Set
language_codewhen the text is not English, so the model does not guess.
Not exposed
These catalog entries describe omitted inputs, pinned values and source constraints. They do not imply that the provider accepts every listed field.
speed— not accepted by this public endpoint.style— not accepted by this public endpoint.pitch— not accepted by this public endpoint.output_format— pinned tomp3_44100_128.seed— not accepted by this public endpoint.
Result
The result kind is audio. Always present: url. Present when the run makes it: timestamps.
{
"data": {
"id": "gen_…",
"kind": "audio",
"model": "elevenlabs-v3",
"status": "completed",
"result": {
"timestamps": [
{}
],
"url": "https://cdn.roasos.com/workspaces/ws_01/result"
}
}
}The files are immutable objects on https://cdn.roasos.com. A result URL is a valid media input for your next request.
Example
The three steps are always the same: quote, run, poll.
# 1. Quote. Nothing is spent.
curl -X POST https://api.roasos.com/v1/audio/elevenlabs-v3 \
-H "Authorization: Bearer roasos_sk_live_..." \
-H "Idempotency-Key: 0f1c9e2a-elevenlabs-v3-0001" \
-H "Content-Type: application/json" \
-d '{"quote_only":true,"text":"Two drops, morning and night. That is the whole routine."}'
# 200 { "data": { "kind": "quote", "quote_id": "q_7Hk...", "credits": 1, "expires_at": "...",
# "confirmation": { "model": "elevenlabs-v3", "example": "1000_characters" } } }
# 2. Run. Same body, same Idempotency-Key, quote_id in place of quote_only.
curl -X POST https://api.roasos.com/v1/audio/elevenlabs-v3 \
-H "Authorization: Bearer roasos_sk_live_..." \
-H "Idempotency-Key: 0f1c9e2a-elevenlabs-v3-0001" \
-H "Content-Type: application/json" \
-d '{"quote_id":"q_7Hk...","text":"Two drops, morning and night. That is the whole routine."}'
# 201 { "data": { "id": "gen_...", "status": "queued", "credits_reserved": 1 } }
# 3. Poll, then read the result.
curl https://api.roasos.com/v1/jobs/gen_.../status -H "Authorization: Bearer roasos_sk_live_..."
curl https://api.roasos.com/v1/jobs/gen_.../result -H "Authorization: Bearer roasos_sk_live_..."The deprecated path
POST /v1/generations still reaches this model for one minor version. It is deprecated. This is how it maps:
POST /v1/generations with { "type": "audio", "model": "elevenlabs-v3", "prompt": "<text>", "options": { "voice": "Charlotte", "stability": 0.4, "language": "en" } }. Today the script goes in prompt, not text. POST /v1/audio/elevenlabs-v3 takes similarity_boost (0 to 1) and apply_text_normalization (auto, on, off). options.speed must be 1 and options.pitch must be 0. timestamps is not returned yet.
See also
llms.txtfor this model — the same facts as one plain-text file for an agent.GET https://api.roasos.com/v1/catalog/elevenlabs-v3— the machine-readable form: enums, defaults, limits and prices. Refer to the API reference.- Jobs — how to poll a job and read its result.
- Provider endpoints:
fal-ai/elevenlabs/tts/eleven-v4.
ElevenLabs Sound Effects v2
Sound effects from a text description: transitions, interface sounds, foley, ambience and seamless loops for ads. The prompt describes the sound. The model never speaks it.
Gemini 3.1 Flash TTS
Text-to-speech on Google Gemini 3.1 Flash TTS, with 30 preset voices and 87 languages. It is best at a voice-over that you direct in plain words: tone, pace, accent, and emotion. It is the lowest per-character voice price in the catalog.