> ## 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.

# HeyGenAvatar6ReferenceToVideoNode - ComfyUI Built-in Node Documentation

> Generate a talking video with HeyGen Avatar 6 from the audio you connect, using a photo or a video as the reference.

Generate a talking video with HeyGen Avatar 6 from the audio you connect, using a photo or a video as the reference. Avatar 6 rebuilds the face and the scene from that reference, so the same audio can drive a still photo, an existing clip, or both.

## Inputs

### Common Inputs

| Parameter | Description | Data Type | Required | Range |
| - | - | - | - | - |
| `model` | Model selector; selecting a model reveals its inputs below. `"heygen-avatar-6-turbo"` is the cheaper, faster tier. | DYNAMIC\_COMBO | Yes | `"heygen-avatar-6"` `"heygen-avatar-6-turbo"` |

### `heygen-avatar-6` Inputs

`"heygen-avatar-6-turbo"` exposes the same inputs with the same limits.

| Parameter | Description | Data Type | Required | Range |
| - | - | - | - | - |
| `image` | Optional photo of the person or character to animate. It sets the look and the scene and is not used as the first frame. | IMAGE | No | Single image |
| `video` | Optional video of an animated or stylized character showing how it talks and moves. Longer videos are trimmed to the first 30 seconds. | VIDEO | No | 2 to 30 seconds |
| `audio` | Speech for the avatar. The video length follows the audio. | AUDIO | Yes | 1.5 seconds to 10 minutes |
| `resolution` | Resolution of the output video. Default: `"1080p"`. | COMBO | Yes | `"1080p"` `"2k"` |
| `aspect_ratio` | Aspect ratio of the output video. `"auto"` keeps the shape of the image, or of the video when no image is connected; other values re-frame the scene. Default: `"auto"`. | COMBO | Yes | `"auto"` `"16:9"` `"9:16"` `"1:1"` `"4:5"` `"5:4"` |
| `motion_prompt` | Optional direction for gestures, expression, posture and camera movement. Default: empty string. | STRING | Yes | Multiline text, up to 1200 characters |
| `motion_expressiveness` | `"less"` gives fewer, smaller gestures. The motion prompt takes precedence where the two overlap. Default: `"default"`. | COMBO | Yes | `"less"` `"default"` |
| `seed` | Seed controls whether the node should re-run; results are non-deterministic regardless of seed. Default: 42. | INT | Yes | 0 to 2147483647 |

**Constraints and behavior:**

* Connect an image, a video, or both; with neither, the node raises an error.
* The image input must hold exactly one image, not a batch.
* Reference videos of realistic people are refused. Use a video of an animated or stylized character, or connect a photo of the person instead.
* A reference video must be 2 seconds or longer and no larger than 50 MB.
* Pricing is per second of output video and depends on the model and the resolution.

## Outputs

| Output Name | Description | Data Type |
| - | - | - |
| `video` | The generated talking video, downloaded from HeyGen once rendering finishes. | VIDEO |

> 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/HeyGenAvatar6ReferenceToVideoNode/en.md)

***

**Source fingerprint (SHA-256):** `1a50342827511e355a1d857c34e8bbfc0b74ca7b003dca783b76fcd151413ff8`


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