> ## Documentation Index
> Fetch the complete documentation index at: https://docs.comfy.org/llms.txt
> Use this file to discover all available pages before exploring further.

# HeyGen Voice 1.0 Instant Clone Text to Speech - ComfyUI Built-in Node Documentation

> Clone a voice from a recording of one speaker and speak text in it with HeyGen Voice 1.0.

Clone a voice from a recording of one speaker and speak text in it with HeyGen Voice 1.0. The node uploads the recording, synthesises the speech and returns it as audio, so a single node goes from a voice sample to finished speech.

Selecting a `model` reveals the parameters specific to it: `audio`, `text`, `language`, `expressiveness_boost`, `recording_language`, `similarity`, `remove_background_noise` and `seed`.

## Inputs

| Parameter | Description | Data Type | Required | Range |
| - | - | - | - | - |
| `model` | Model to use. `heygen-voice-1-turbo` is faster and costs less (default: `"heygen-voice-1"`). | DYNAMIC\_COMBO | Yes | `"heygen-voice-1"` `"heygen-voice-1-turbo"` |
| `audio` | Recording of one speaker to clone. Use at least 10 seconds of clear speech; only the first 3 minutes are used. | AUDIO | Yes | - |
| `text` | Text to speak, up to 5000 characters. Speech longer than about 12 minutes fails, so split very long Chinese or Japanese text (default: empty). | STRING | Yes | 1 to 5000 characters |
| `language` | Language of the text. Set it for closely related languages such as Serbian or Malay, which auto-detection can confuse (default: `"auto"`). | COMBO | Yes | `"auto"` Multiple language presets |
| `expressiveness_boost` | Higher values are more expressive; lower values stay closer to the original voice (default: 0.5). | FLOAT | Yes | 0.0 to 1.0 (step 0.05) |
| `recording_language` | Language spoken in the recording, which narrows the clone to that language (default: `"auto"`). | COMBO | Yes | `"auto"` Multiple language presets |
| `similarity` | How closely the voice follows the recording (default: `"max"`). | COMBO | Yes | `"medium"` `"high"` `"extra_high"` `"max"` |
| `remove_background_noise` | Clean up background noise in the recording before cloning (default: True). | BOOLEAN | Yes | `True` `False` |
| `seed` | Seed controls whether the node re-runs; results are non-deterministic regardless of seed (default: 42). | INT | Yes | 0 to 2147483647 |

`text` must contain at least one character and `audio` is required: a recording of several speakers, or one too noisy to clone, gives a poor voice. Set `recording_language` when the language of the sample is not obvious.

## Outputs

| Output Name | Description | Data Type |
| - | - | - |
| `AUDIO` | The text spoken in the cloned voice. | AUDIO |

> 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/HeyGenVoiceCloneTextToSpeechNode/en.md)

***

**Source fingerprint (SHA-256):** `ea18e2abce687e106ed0b37eb7afd723fbe72a3b4d58124a143c00b66d2fe7c9`


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.