API ReferenceText-to-Speech ModelsElevenLabs
Flash v2.5
ElevenLabs Flash v2.5 — ultra low-latency TTS in 32 languages.
ElevenLabs' ultra low-latency model across 32 languages, with the highest character limit (40,000). Best for long scripts, conversational agents, and anything latency-sensitive.
| Endpoint | POST /api/v1/tts/synthesize |
Model ID (ttsModel) | eleven-flash-v2-5 |
| Provider | ElevenLabs (paid access) |
| Character Limit | 40,000 |
| Cost | 1,050 / 1K chars |
Shared conventions
Auth, the voice→provider rule, rate limits, and the full error table live on the Speech API page. Below is only what's specific to Flash v2.5.
Model-specific notes
- Tuned for speed and long form — the 40,000-char ceiling makes it the best fit for articles and chapters.
- Supports
language_override(force the voice's native language) andspeed. - No
style_exaggeration— that field is ignored (it applies only to Multilingual v2).
Request body
| Parameter | Type | Required | Description |
|---|---|---|---|
text | string | Yes | Text to synthesize (max 40,000 chars) |
voice | string | Yes | An ElevenLabs voice ID from GET /api/v1/tts/voices |
ttsModel | string | Yes | Must be "eleven-flash-v2-5" |
speed | number | No | Playback speed · Default: 1.0 |
language_override | boolean | No | Force the voice's native language · Default: false |
stability | number | No | Voice stability 0.0–1.0 · Default: 0.5 |
similarity_boost | number | No | Similarity to source voice 0.0–1.0 · Default: 0.75 |
speaker_boost | boolean | No | Enhance speaker clarity · Default: true |
response_format | string | No | "mp3_44100_128" "pcm_8000" "pcm_16000" "pcm_22050" "pcm_24000" · Default: "mp3_44100_128" |
title | string | No | Title for the saved project |
{
"text": "Your order has shipped and will arrive within two business days.",
"voice": "db4e815d-00aa-43e6-99cf-0d9b4db9a07a",
"ttsModel": "eleven-flash-v2-5"
}Response (200 OK)
{
"url": "https://cdn.sonnalabs.app/sonna/api-ephemeral/tts/paid/user123/ghi789.mp3",
"remainingCredits": 99895,
"projectCreated": true
}Errors (400 TEXT_TOO_LONG over 40,000 chars, 403 PAID_ACCESS_REQUIRED, 409, 429, 503) follow the shared Speech API table.