ElevenLabs Block

The ElevenLabs block generates realistic speech from text using ElevenLabs voices. Reach for it whenever a workflow needs to turn text into natural-sounding audio output.

Overview

When to Use

• Convert workflow-generated text (summaries, responses, narration) into spoken audio.
• Add voiceover to automated content pipelines.
• Produce audio prompts or notifications from dynamic text.
• Generate multilingual speech using ElevenLabs multilingual or turbo models.
• Create audio assets on the fly from upstream block outputs.

Configuration

Text

The text to convert to speech. Required. Accepts static text or a reference to an upstream output.

Voice ID

The ElevenLabs voice identifier to use for synthesis. Required.

Model ID

Dropdown to select the synthesis model. Options: eleven_monolingual_v1, eleven_multilingual_v2, eleven_turbo_v2, eleven_turbo_v2_5, eleven_flash_v2_5.

API Key

Your ElevenLabs API key, stored as a password field. Required.

Inputs & Outputs

• Inputs:
  • text (string) — Text to convert to speech
  • voiceId (string) — Voice identifier
  • modelId (string) — Model identifier
  • apiKey (string) — ElevenLabs API key
• Outputs:
  • audioUrl (string) — Generated audio URL

Tools

ElevenLabs TTS (elevenlabs_tts) — Sends the text, voice ID, model ID, and API key to the ElevenLabs API via a proxy and returns a URL pointing to the generated audio file. The modelId defaults to eleven_monolingual_v1 when not specified.

YAML Example

elevenlabs_1:
  type: elevenlabs
  name: "ElevenLabs"
  inputs:
    text: "{{start.input}}"
    voiceId: "21m00Tcm4TlvDq8ikWAM"
    modelId: "eleven_multilingual_v2"
    apiKey: "{{ELEVENLABS_API_KEY}}"
  connections:
    outgoing:
      - target: next-block-id