ElevenLabs is a state-of-the-art text-to-speech platform that creates incredibly natural and expressive AI voices. It offers some of the most realistic and emotionally nuanced synthetic voices available today, making it ideal for creating lifelike audio content.
With ElevenLabs, you can:
- Generate natural-sounding speech: Create audio that's nearly indistinguishable from human speech
- Choose from diverse voice options: Access a library of pre-made voices with different accents, tones, and characteristics
- Clone voices: Create custom voices based on audio samples (with proper permissions)
- Control speech parameters: Adjust stability, clarity, and emotional tone to fine-tune output
- Add realistic emotions: Incorporate natural-sounding emotions like happiness, sadness, or excitement
In Sim, the ElevenLabs integration enables your agents to convert text to lifelike speech, enhancing the interactivity and engagement of your applications. This is particularly valuable for creating voice assistants, generating audio content, developing accessible applications, or building conversational interfaces that feel more human. The integration allows you to seamlessly incorporate ElevenLabs' advanced speech synthesis capabilities into your agent workflows, bridging the gap between text-based AI and natural human communication.
Usage Instructions
Integrate ElevenLabs into the workflow. Can convert text to speech.
Tools
elevenlabs_tts
Convert TTS using ElevenLabs voices
Input
| Parameter | Type | Required | Description |
|---|---|---|---|
text | string | Yes | The text to convert to speech (e.g., "Hello, welcome to our service!") |
voiceId | string | Yes | The ID of the voice to use (e.g., "21m00Tcm4TlvDq8ikWAM" for Rachel) |
modelId | string | No | The ID of the model to use (e.g., "eleven_multilingual_v2", "eleven_turbo_v2"). Defaults to eleven_monolingual_v1 |
stability | number | No | Voice stability setting from 0.0 to 1.0 (e.g., 0.5 for balanced, 0.75 for more stable). Higher values produce more consistent output |
similarity | number | No | Similarity boost setting from 0.0 to 1.0 (e.g., 0.75 for natural, 1.0 for maximum similarity). Higher values make the voice more similar to the original |
apiKey | string | Yes | Your ElevenLabs API key |
Output
| Parameter | Type | Description |
|---|---|---|
audioUrl | string | The URL of the generated audio |
audioFile | file | The generated audio file |