Video

Uploads videos to the Vimmerse platform for use in API requests.

Benefits

• Store videos securely in the cloud
• Reference videos by URL in other API calls
• Access videos from anywhere
• Manage your media assets efficiently

Supported Formats

• MP4 (recommended)
• MOV
• AVI
• WebM

Usage

After uploading, you'll receive a URL that can be used with the video_url parameter in other video processing endpoints.

Changes the aspect ratio of a video by cropping or reframing.

How It Works

• Crops the video to match the target aspect ratio
• Maintains video quality and content
• Automatically centers content or allows custom positioning

Parameters

• video_file or video_url: Input video (required)
• aspect_ratio: Desired aspect ratio (required)
  • Available: '21:9', '16:9', '4:3', '1:1', '3:4', '9:16', '9:21'
• offset_x: Horizontal offset from top-left corner (optional)
• offset_y: Vertical offset from top-left corner (optional)

Use Cases

• Convert between landscape and portrait formats
• Prepare videos for different social media platforms
• Adapt content for different screen sizes

Example

import requests

url = "https://api.vimmerse.net/video/crop-to-aspect-ratio"

files=[
    ('video_file', ('video.mp4', open('/path/to/video.mp4', 'rb'), 'video/mp4'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    "aspect_ratio": "9:16",  # Vertical format for TikTok, Instagram Stories
    "offset_x": 0,  # Optional: horizontal crop offset
    "offset_y": 0   # Optional: vertical crop offset
}

response = requests.request("POST", url, headers=headers, data=payload, files=files)
print(response.text)

Reframes video using AI to reposition and resize content while maintaining quality.

How It Works

Luma AI's reframing technology intelligently adjusts camera position to fit different aspect ratios without losing important content.

Parameters

• video_file or video_url: Input video (required)
• aspect_ratio: Target aspect ratio (required)
  • Available: '21:9', '16:9', '4:3', '1:1', '3:4', '9:16', '9:21'
• prompt: Optional guidance for reframing behavior
• grid_position_x: X grid position (optional)
• grid_position_y: Y grid position (optional)
• x_start, x_end: Horizontal start/end positions (optional)
• y_start, y_end: Vertical start/end positions (optional)

Best Practices

• Works best with clear foreground subjects
• Automatic centering by default
• Custom positioning available for precise control

Example

import requests

url = "https://api.vimmerse.net/video/reframe"

files=[
    ('video_file', ('video.mp4', open('/path/to/video.mp4', 'rb'), 'video/mp4'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    "aspect_ratio": "16:9",
    "prompt": "Keep the main subject centered and visible"
}

response = requests.request("POST", url, headers=headers, data=payload, files= precinct)
print(response.text)

Enhances video quality using AI-powered upscaling and noise reduction.

Enhancement Options

4X

• Upscales video to 4x resolution
• Maintains sharp details
• Best for: Large screen displays, high-resolution exports

2X

• Upscales video to 2x resolution
• Faster processing than 4X
• Best for: General quality improvement

4X&DeepClean

• Combines 4X upscaling with noise reduction
• Removes artifacts and imperfections
• Best for: Old or low-quality footage

2X&DeepClean

• Combines 2X upscaling with noise reduction
• Balanced quality and speed
• Best for: Moderate quality improvements

DeepClean

• Noise reduction only, no upscaling
• Removes grain and artifacts
• Best for: Already high-resolution videos with noise

Parameters

• video_file or video_url: Input video (required)
• enhance_mode: Enhancement option (required)
  • Available: '4X', '2X', '4X&DeepClean', '2X&DeepClean', 'DeepClean'
• compression_scale: Output compression, 0-1 (default: '0.5')

Example

import requests

url = "https://api.vimmerse.net/video/enhance"

files=[
    ('video_file', ('video.mp4', open('/path/to/video.mp4', 'rb'), 'video/mp4'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    'enhance_mode': '4X&DeepClean',  # Maximum quality enhancement
    'compression_scale': '0.7'       # Higher quality, larger file
}

response = requests.request("POST", url, headers=headers, data=payload, files=files)
print(response.text)

Adds AI-generated sound effects to silent or noisy videos.

How It Works

• Analyzes video content to understand the scene
• Generates appropriate sound effects based on prompt
• Syncs audio with visual elements
• Enhances viewer immersion

Parameters

• video_file or video_url: Input video (required)
• prompt: Description of desired sound effects (optional but recommended)
  • Examples: "crowd cheering", "ocean waves", "car engine", "footsteps on gravel"

Use Cases

• Add ambiance to silent videos
• Enhance product demonstrations
• Create ASMR-style content
• Improve video engagement

Example

import requests

url = "https://api.vimmerse.net/video/sound-effect"

files=[
    ('video_file', ('video.mp4', open('/path/to/video.mp4', 'rb'), 'video/mp4'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    'prompt': 'Add vibrant cafe ambience with coffee brewing and soft chatter'
}

response = requests.request("POST", url, headers=headers, data=payload, files=files)
print(response.text)

Creates lip-sync videos using pre-designed avatars with customizable voices.

How It Works

• Select from a library of professional avatars
• Choose from multiple voice options
• Add background images and music
• Generate realistic talking videos

Parameters

• avatar_id: Avatar ID (required)
  • Get available avatars from: https://api.vimmerse.net/video/avatars/avatars
• voice_id: Voice ID (required)
  • Get available voices from: https://api.vimmerse.net/video/avatars/voices
• prompt: Speech text (required)
• bg_image_file or bg_image_url: Background image (optional)
• audio_file or audio_url: Background music (optional)
• aspect_ratio: Video format - '1:1', '16:9', '9:16' (default: '16:9')
• show_caption: Enable captions - 'On' or 'Off' (default: 'On')

Use Cases

• Create professional spokesperson videos
• Generate educational content
• Create marketing videos
• Produce accessible content with captions

Example

import requests

url = "https://api.vimmerse.net/video/avatars/lip-sync"

headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    'avatar_id': 'avatar_123',  # Get from avatars endpoint
    'voice_id': 'voice_456',    # Get from voices endpoint
    'prompt': 'Introducing our new product line',
    'bg_image_url': 'https://example.com/background.jpg',
    'aspect_ratio': '9:16',
    'show_caption': 'On'
}

response = requests.request("POST", url, headers=headers, data=payload)
print(response.text)

Generates lip-sync videos by making faces speak audio or text prompts.

How It Works

• Accepts image, video, audio, or text as input
• Synchronizes lip movement with speech
• Generates realistic talking video

Input Requirements

You need at least one visual input AND one audio input:

Visual Input:

• image_file or image_url: Static image of a person
• video_file or video_url: Video with a person

Audio Input:

• audio_file or audio_url: Existing audio file
• prompt + voice: Text-to-speech conversion

Parameters

• prompt: Speech text (required if no audio)
• voice: Voice selection (required if using prompt)
  • Available: "Aria", "Roger", "Sarah", "Laura", "Charlie", "George", "Callum", "River", "Liam", "Charlotte", "Alice", "Matilda", "Will", "Jessica", "Eric", "Chris", "Brian", "Daniel", "Lily", "Bill"
• option: Lip-sync method
  • "Auto": Automatically selects best method
  • "Sync", "HummingBird": For video input
  • "Hedra", "OminHuman": For image input

Use Cases

• Create talking avatars
• Add voiceovers to videos
• Generate spokesperson videos
• Create personalized messages

Example

import requests

url = "https://api.vimmerse.net/video/lip-sync"

files=[
    ('image_file', ('person.jpg', open('/path/to/person.jpg', 'rb'), 'image/jpeg'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    'prompt': 'Welcome to our product launch event!',
    'voice': 'Sarah',
    'option': 'Auto'
}

response = requests.request("POST", url, headers=headers, data=payload, files=files)
print(response.text)

Generates video effects by animating images with specified effects.

Supported Effects and Tools

PikaEffects

Available effects: Explode, Zoom in, Zoom out, Pan left, Pan right, etc.

KlingEffects

Creative animation effects for images.

WanEffects

AI-powered video effects with aspect ratio control.

ViduEffects

Video generation effects with temporal consistency.

PixverseEffects

Style-preserving video effects.

Parameters

• image_file or image_url: Base image (required)
• ref_image_file or ref_image_url: Additional image for specific effects (optional)
• effect: Description of desired effect (required)
  • Examples: "explode", "zoom in", "rotate", "fade", etc.
• option: Effect tool (required)
  • Available: "PikaEffects", "KlingEffects", "WanEffects", "ViduEffects", "PixverseEffects"
• aspect_ratio: Video aspect ratio (optional)
• quantity: Number of variations (1-4, default: 1)

Example

import requests

url = "https://api.vimmerse.net/video/video-effects"

files=[
    ('image_file', ('image.png', open('/path/to/image.png', 'rb'), 'image/png'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    'effect': 'zoom out slowly while rotating',
    'option': 'PikaEffects',
    'aspect_ratio': '16:9',
    'quantity': 2
}

response = requests.request("POST", url, headers=headers, data=payload, files=files)
print(response.text)

Replaces the background of a video with another video or image.

How It Works

• Extracts the foreground object from the video using advanced AI
• Replaces the background with your chosen video or image
• Supports chroma key (green screen) masking for precise control

Parameters

• video_file or video_url: Input video with background to replace (required)
• bg_video_file or bg_video_url: Background video (optional)
• bg_image_file or bg_image_url: Background image (optional)
• chroma_key_color: HTML color code for green screen removal (default: 'origin')
  • Use 'origin' for regular foreground masking
  • Use HTML codes like '#00FF00' for green, '#FFFFFF' for white
• chroma_tolerance: Color matching tolerance, 0-1 (default: '0.1')
• overlay_scale: Scale of foreground, 0-1 (default: '0.6666')
• alignment: Foreground alignment - 'center', 'left', 'right' (default: 'center')
• start_time: Start timestamp for video background (default: 0)
• crop_to_content: Crop video to content (default: True)
• x_offset: Horizontal offset for foreground (default: 0)
• y_offset: Vertical offset for foreground (default: 0)

Use Cases

• Product showcase videos
• Remove studio backgrounds
• Add dynamic video backgrounds
• Green screen compositing

Example: Video Background Replacement

import requests

url = "https://api.vimmerse.net/video/replace-background"

payload = {
    'video_url': 'https://example.com/main_video.mp4',
    'bg_video_url': 'https://example.com/background_video.mp4',
    'chroma_key_color': 'origin',  # or '#00FF00' for green screen
    'alignment': 'center'
}

headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

response = requests.request("POST", url, headers=headers, data=payload)
print(response.text)

Example: Image Background Replacement

import requests

url = "https://api.vimmerse.net/video/replace-background"

payload = {
    'video_url': 'https://example.com/main_video.mp4',
    'bg_image_url': 'https://example.com/background_image.jpg'
}

headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

response = requests.request("POST", url, headers=headers, data=payload)
print(response.text)

Generates videos by animating multiple element images with prompts.

How It Works

• Upload multiple images representing elements
• Provide a prompt describing the desired animation
• AI generates a cohesive video with all elements

Parameters

• image_files or image_urls: List of element images (required)
• prompt: Description of desired video (required)
• aspect_ratio: Output video ratio (optional)
• webhook_url: URL to receive completion notification (optional)

Use Cases

• Product showcases with multiple items
• Character animation
• Multi-element storytelling
• Interactive demonstrations

Example

import requests

url = "https://api.vimmerse.net/video/reference-to-video"

files=[
    ('image_files', ('character1.png', open('/path/to/char1.png', 'rb'), 'image/png')),
    ('image_files', ('character2.png', open('/path/to/char2.png', 'rb'), 'image/png'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    'prompt': 'Two friends walking together through a sunny park',
    'aspect_ratio': '16:9'
}

response = requests.request("POST", url, headers=headers, data=payload, files=files)
print(response.text)

Modifies existing videos by applying changes guided by prompts and reference images.

How It Works

• Uses Luma AI's video editing technology
• Requires both video and image inputs
• Applies changes based on text prompts
• Controls creativity level for editing strength

Parameters

• video_file or video_url: Input video to modify (required)
• image_file or image_url: Reference image for modifications (required)
• prompt: Description of desired modifications (required)
  • Examples: "Make the character jump higher", "Change the sky to sunset", "Add more motion"
• creativity_strength: Editing intensity, 1-10 (default: 4)
  • Lower values: Stay closer to original
  • Higher values: More creative changes
• option: Modification method (default: "LumaModify")

Use Cases

• Fix or enhance video content
• Apply style changes
• Modify specific elements
• Create video variations

Example

import requests

url = "https://api.vimmerse.net/video/modify"

files=[
    ('video_file', ('video.mp4', open('/path/to/video.mp4', 'rb'), 'video/mp4')),
    ('image_file', ('reference.jpg', open('/path/to/reference.jpg', 'rb'), 'image/jpeg'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    'prompt': 'Make the character jump higher and add more energy',
    'creativity_strength': 6
}

response = requests.request("POST", url, headers=headers, data=payload, files=files)
print(response.text)

Attaches audio tracks to videos with multiple synchronization options.

Attachment Modes

MatchVideo

• Trims or loops audio to match video length
• Best for: Background music that needs to fit video duration

PadAudio

• Pads audio with silence to match video length
• Best for: Short audio that needs to cover longer videos

MatchAudio

• Trims video to match audio length
• Best for: Keeping full audio and adjusting video

SpeedAdjust

• Adjusts video playback speed to match audio duration
• Best for: Dramatic effect with speed changes

Strict

• Requires audio and video to have the same duration
• Best for: Exact synchronization needs

Parameters

• video_file or video_url: Input video (required)
• audio_file or audio_url: Audio file to attach (required)
• attach_mode: Synchronization method (required)
• webhook_url: Completion notification URL (optional)

Example

import requests

url = "https://api.vimmerse.net/video/audio"

files=[
    ('video_file', ('video.mp4', open('/path/to/video.mp4', 'rb'), 'video/mp4')),
    ('audio_file', ('music.mp3', open('/path/to/music.mp3', 'rb'), 'audio/mp3'))
]
headers = {
    'X-Api-Key': 'YOUR_API_KEY'
}

payload = {
    'attach_mode': 'MatchVideo',  # Audio will loop to match video length
}

response = requests.request("POST", url, headers=headers, data=payload, files=files)
print(response.text)

Generate video from text prompt and optional images. Supports various AI models including LumaAI, RunwayML, KlingAI, Veo, Sora, and more.