> ## Documentation Index
> Fetch the complete documentation index at: https://docs.pinkfish.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# video-processing
> AI video generation and audio extraction
**Server path:** `/video-processing` | **Type:** Embedded | **PCID required:** No
## Tools
| Tool | Description |
| ------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| [`video-processing_veo_generate`](#video-processing_veo_generate) | Start AI video generation from text prompts using Google VEO models. Returns a prediction ID immediately - use video-processing\_veo\_poll to check status and get the video URL when complete. Video generation typically takes 1-2 minutes. |
| [`video-processing_veo_poll`](#video-processing_veo_poll) | Check the status of a VEO video generation job. Call this after video-processing\_veo\_generate to check if the video is ready. Poll every 10-15 seconds until status is "succeeded" or "failed". |
| [`video-processing_audio_extract`](#video-processing_audio_extract) | Extract audio from video files. Supports both synchronous (for small files) and asynchronous (for large files) processing. Output is MP3 format. |
***
## video-processing\_veo\_generate
Start AI video generation from text prompts using Google VEO models. Returns a prediction ID immediately - use video-processing\_veo\_poll to check status and get the video URL when complete. Video generation typically takes 1-2 minutes.
**Parameters:**
| Parameter | Type | Required | Default | Description |
| ------------------ | --------- | -------- | ---------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| `prompt` | string | Yes | — | Required: Text description of the video to generate. Be specific and descriptive for best results. |
| `model` | string | No | `"veo-3.1-fast"` | Model to use. "veo-3.1" for full-featured high quality, "veo-3.1-fast" for speed-optimized (default). Note: reference\_images only work with veo-3.1. |
| `duration` | number | No | `8` | Video duration in seconds (default: 8) |
| `aspect_ratio` | string | No | `"16:9"` | Aspect ratio. "16:9" for landscape/YouTube (default), "9:16" for vertical/mobile/TikTok |
| `resolution` | string | No | `"1080p"` | Video resolution (default: 1080p) |
| `image` | string | No | — | Optional: Starting image URL (1280x720 or 720x1280) to animate into video |
| `last_frame` | string | No | — | Optional: Ending image URL for smooth transition. Requires "image" to be set. |
| `reference_images` | string\[] | No | — | Optional: Array of 1-3 image URLs for style guidance (e.g., \["[https://example.com/ref1.jpg](https://example.com/ref1.jpg)", "[https://example.com/ref2.jpg](https://example.com/ref2.jpg)"]). Only works with veo-3.1 model, 16:9 aspect ratio, and 8-second duration. |
| `generate_audio` | boolean | No | `true` | Whether to generate audio for the video (default: true) |
| `negative_prompt` | string | No | — | Optional: What to avoid in the video (e.g., "blurry, low quality") |
| `seed` | number | No | — | Optional: Seed for reproducible results |
```json theme={null}
{
"type": "object",
"properties": {
"prompt": {
"type": "string",
"description": "Required: Text description of the video to generate. Be specific and descriptive for best results."
},
"model": {
"type": "string",
"enum": [
"veo-3.1",
"veo-3.1-fast"
],
"default": "veo-3.1-fast",
"description": "Model to use. \"veo-3.1\" for full-featured high quality, \"veo-3.1-fast\" for speed-optimized (default). Note: reference_images only work with veo-3.1."
},
"duration": {
"type": "number",
"default": 8,
"description": "Video duration in seconds (default: 8)"
},
"aspect_ratio": {
"type": "string",
"enum": [
"16:9",
"9:16"
],
"default": "16:9",
"description": "Aspect ratio. \"16:9\" for landscape/YouTube (default), \"9:16\" for vertical/mobile/TikTok"
},
"resolution": {
"type": "string",
"enum": [
"1080p",
"720p"
],
"default": "1080p",
"description": "Video resolution (default: 1080p)"
},
"image": {
"type": "string",
"description": "Optional: Starting image URL (1280x720 or 720x1280) to animate into video"
},
"last_frame": {
"type": "string",
"description": "Optional: Ending image URL for smooth transition. Requires \"image\" to be set."
},
"reference_images": {
"type": "array",
"items": {
"type": "string"
},
"description": "Optional: Array of 1-3 image URLs for style guidance (e.g., [\"https://example.com/ref1.jpg\", \"https://example.com/ref2.jpg\"]). Only works with veo-3.1 model, 16:9 aspect ratio, and 8-second duration."
},
"generate_audio": {
"type": "boolean",
"default": true,
"description": "Whether to generate audio for the video (default: true)"
},
"negative_prompt": {
"type": "string",
"description": "Optional: What to avoid in the video (e.g., \"blurry, low quality\")"
},
"seed": {
"type": "number",
"description": "Optional: Seed for reproducible results"
}
},
"required": [
"prompt"
]
}
```
***
## video-processing\_veo\_poll
Check the status of a VEO video generation job. Call this after video-processing\_veo\_generate to check if the video is ready. Poll every 10-15 seconds until status is "succeeded" or "failed".
**Parameters:**
| Parameter | Type | Required | Default | Description |
| -------------- | ------ | -------- | ------- | ---------------------------------------------------------------------------------------------- |
| `predictionId` | string | Yes | — | Required: The prediction ID returned by video-processing\_veo\_generate (e.g., "pred\_abc123") |
```json theme={null}
{
"type": "object",
"properties": {
"predictionId": {
"type": "string",
"description": "Required: The prediction ID returned by video-processing_veo_generate (e.g., \"pred_abc123\")"
}
},
"required": [
"predictionId"
]
}
```
***
## video-processing\_audio\_extract
Extract audio from video files. Supports both synchronous (for small files) and asynchronous (for large files) processing. Output is MP3 format.
**Parameters:**
| Parameter | Type | Required | Default | Description |
| --------------------------- | ------- | -------- | ------- | ------------------------------------------------------------------------------------------------------------- |
| `fileUrl` | string | Yes | — | Required: URL of the video file to extract audio from |
| `collectionId` | string | No | — | Optional: Filestorage collection ID to save the extracted audio to. If not provided, uses default collection. |
| `outputFileName` | string | No | — | Optional: Custom filename for the extracted audio (e.g., "my\_audio.mp3") |
| `async` | boolean | No | `false` | Use async processing for large files (default: false). Async returns immediately with a job ID. |
| `file_links_expire_in_days` | number | No | `7` | Number of days before the download link expires (default: 7) |
```json theme={null}
{
"type": "object",
"properties": {
"fileUrl": {
"type": "string",
"description": "Required: URL of the video file to extract audio from"
},
"collectionId": {
"type": "string",
"description": "Optional: Filestorage collection ID to save the extracted audio to. If not provided, uses default collection."
},
"outputFileName": {
"type": "string",
"description": "Optional: Custom filename for the extracted audio (e.g., \"my_audio.mp3\")"
},
"async": {
"type": "boolean",
"default": false,
"description": "Use async processing for large files (default: false). Async returns immediately with a job ID."
},
"file_links_expire_in_days": {
"type": "number",
"default": 7,
"description": "Number of days before the download link expires (default: 7)"
}
},
"required": [
"fileUrl"
]
}
```