The ElevenLabs Instant Voice Clone node creates a new, unique voice model by analyzing 1 to 8 audio recordings of a person's voice. It sends these samples to the ElevenLabs API, which processes them to generate a voice clone that can be used for text-to-speech synthesis.

## Inputs

| Parameter | Description | Data Type | Required | Range |
| --- | --- | --- | --- | --- |
| `audio_*` | Audio recordings for voice cloning. You must provide between 1 and 8 audio files. | AUDIO | Yes | 1 to 8 files |
| `remove_background_noise` | Remove background noise from voice samples using audio isolation. (default: False) | BOOLEAN | No | True / False |

**Note:** You must provide at least one audio file, and you can provide up to eight. The node will automatically create input slots for the audio files you add.

## Outputs

| Output Name | Description | Data Type |
| --- | --- | --- |
| `voice` | The unique identifier for the newly created cloned voice model. This output can be connected to other ElevenLabs text-to-speech nodes. | ELEVENLABS_VOICE |

> This documentation was AI-generated. If you find any errors or have suggestions for improvement, please feel free to contribute! [Edit on GitHub](https://github.com/Comfy-Org/embedded-docs/blob/main/comfyui_embedded_docs/docs/ElevenLabsInstantVoiceClone/en.md)

---
**Source fingerprint (SHA-256):** `760bd99ed8382eb875ab21739346d178130b02240526308b784332a59b048d78`