Sonna
API ReferenceText-to-Speech ModelsElevenLabs

Multilingual v2

ElevenLabs Multilingual v2 — life-like, emotionally rich voices in 29 languages.

ElevenLabs' most life-like, emotionally rich model across 29 languages. The go-to for audiobooks, voice-overs, and Professional Voice Clones where consistency matters more than latency.

EndpointPOST /api/v1/tts/synthesize
Model ID (ttsModel)eleven-multilingual-v2
ProviderElevenLabs (paid access)
Character Limit10,000
Cost2,100 / 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 Multilingual v2.

Model-specific notes

  • The only model with a style_exaggeration slider — use it to push expressiveness (v3 uses audio tags instead; Flash has none).
  • Honors a speed setting.
  • No language_override — Multilingual v2 doesn't take a language code; it infers language from the text.
  • Best similarity for Professional Voice Clones (per ElevenLabs guidance).

Request body

ParameterTypeRequiredDescription
textstringYesText to synthesize (max 10,000 chars)
voicestringYesAn ElevenLabs voice ID from GET /api/v1/tts/voices
ttsModelstringYesMust be "eleven-multilingual-v2"
style_exaggerationnumberNoExpressiveness 0.01.0 · Default: 0.0
speednumberNoPlayback speed · Default: 1.0
stabilitynumberNoVoice stability 0.01.0 · Default: 0.5
similarity_boostnumberNoSimilarity to source voice 0.01.0 · Default: 0.75
speaker_boostbooleanNoEnhance speaker clarity · Default: true
response_formatstringNo"mp3_44100_128" "pcm_8000" "pcm_16000" "pcm_22050" "pcm_24000" · Default: "mp3_44100_128"
titlestringNoTitle for the saved project
{
  "text": "Selamat datang. Audio narasi Anda telah selesai dibuat.",
  "voice": "db4e815d-00aa-43e6-99cf-0d9b4db9a07a",
  "ttsModel": "eleven-multilingual-v2",
  "stability": 0.6,
  "style_exaggeration": 0.3
}

Response (200 OK)

{
  "url": "https://cdn.sonnalabs.app/sonna/api-ephemeral/tts/paid/user123/def456.mp3",
  "remainingCredits": 98950,
  "projectCreated": true
}

Errors (400 TEXT_TOO_LONG over 10,000 chars, 403 PAID_ACCESS_REQUIRED, 409, 429, 503) follow the shared Speech API table.

On this page