version: "1.0.0" name: elevenlabs-tts description: "Text-to-speech with ElevenLabs: generate realistic voice audio from text. Requires ELEVENLABS_API_KEY." requires_env:

• ELEVENLABS_API_KEY

ElevenLabs Text-to-Speech

Generate realistic voice audio from text.

When to Use

• "Read this aloud"
• "Generate audio of this text"
• "Create a voice message"
• Any text-to-speech request

List Available Voices

curl -s "https://api.elevenlabs.io/v1/voices" \
  -H "xi-api-key: $ELEVENLABS_API_KEY" | jq '.voices[] | {name: .name, voice_id: .voice_id}'

Generate Speech

curl -s -X POST "https://api.elevenlabs.io/v1/text-to-speech/VOICE_ID" \
  -H "xi-api-key: $ELEVENLABS_API_KEY" \
  -H "Content-Type: application/json" \
  --output speech.mp3 \
  -d '{
    "text": "Text to convert to speech",
    "model_id": "eleven_multilingual_v2",
    "voice_settings": {
      "stability": 0.5,
      "similarity_boost": 0.75
    }
  }'

Default Voice IDs

• 21m00Tcm4TlvDq8ikWAM — Rachel (female, calm)
• EXAVITQu4vr4xnSDxMaL — Bella (female, warm)
• ErXwobaYiN019PkySvjV — Antoni (male, articulate)

Use the List Voices endpoint to find all available voices.

Models

• eleven_multilingual_v2 — Best quality, 29 languages
• eleven_monolingual_v1 — English only, fast
• eleven_turbo_v2 — Lowest latency

Notes

• Free tier: 10,000 characters/month
• Output: mp3
• After generating, deliver the audio file to the user