> ## 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.

# Pruna P-Video-2 Pro Avatar - ComfyUI Built-in Node Documentation

> Turn a portrait into a talking video with Pruna P-Video-2 Pro Avatar.

Turn a portrait into a talking video with Pruna P-Video-2 Pro Avatar. Speak a script through text-to-speech, or connect your own audio and the person lip-syncs it.

## Inputs

| Parameter | Description | Data Type | Required | Range |
| - | - | - | - | - |
| `image` | Portrait the video starts from. Aspect ratios from 1:2 to 4:1 are kept; more extreme images, such as 1:5, are cropped to 9:16 or 16:9. | IMAGE | Yes | - |
| `speech` | Drive the video with a text-to-speech script or with your own audio. Selecting an option reveals its inputs below. | DYNAMIC\_COMBO | Yes | `"script"` `"audio"` |
| `prompt` | Describes what the video shows (default: "The person is talking."). Up to 5000 characters. | STRING | Yes | Up to 5000 characters |
| `resolution` | Output resolution (default: `"768p"`). 1080p, 2k and 4k upscale the generated 768p video. | COMBO | Yes | `"480p"` `"768p"` `"1080p"` `"2k"` `"4k"` |
| `enhance_prompt` | Add the motion seen in the portrait, such as background, lights and body movement, to the prompt (default: True). Advanced setting. | BOOLEAN | Yes | `True` `False` |
| `seed` | Seed for the generation (default: 42). The same seed reproduces a result exactly only when `enhance_prompt` is off. | INT | Yes | 0 to 2147483647 Step: 1 |

### `"script"` Inputs

| Parameter | Description | Data Type | Required | Range |
| - | - | - | - | - |
| `speech.script` | Words the person speaks. The video matches the length of the speech, and speech longer than 3 minutes is cut at 3 minutes. | STRING | Yes | Any text |
| `speech.voice` | Voice used for the speech (default: `"Zephyr (Female)"`). | COMBO | Yes | 30 preset voices |
| `speech.language` | Language the script is spoken in (default: `"English (US)"`). | COMBO | Yes | `"English (US)"` `"English (UK)"` `"Spanish"` `"French"` `"German"` `"Italian"` `"Portuguese (Brazil)"` `"Japanese"` `"Korean"` `"Hindi"` |
| `speech.voice_prompt` | How the script is spoken: tone, pacing, accent, emotion (default: "Say the following."). It is not spoken itself. | STRING | Yes | Any text |

### `"audio"` Inputs

| Parameter | Description | Data Type | Required | Range |
| - | - | - | - | - |
| `speech.audio` | Speech the person lip-syncs. The video matches its length. | AUDIO | Yes | At least 1 second |

**Notes:**

* Longer output limits the audio length: 2k accepts up to 150 seconds of audio and 4k up to 60 seconds. Longer audio at those resolutions stops the node with an error, so use a lower resolution or shorter clip.
* The portrait aspect ratio fixes the output shape. Very tall or very wide portraits are cropped rather than padded, so keep the subject clear of the edges.
* With `enhance_prompt` off the prompt is used as written, which is what makes the same seed reproduce a result exactly.

## Outputs

| Output Name | Description | Data Type |
| - | - | - |
| `video` | The talking video generated from the portrait. | 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/PrunaAvatarNode/en.md)

***

**Source fingerprint (SHA-256):** `58efc7a110734f8da0a771371c5230908bf8226a946d69c5315e51e259630043`


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