ComfyUI-MediaUtilities

ComfyUI-MediaUtilities

Custom nodes for loading and previewing media from URLs in ComfyUI.

Features

• Load Audio From URL: Download and load audio files from the internet
• Preview Audio: Play downloaded audio in the ComfyUI interface
• Load Video From URL: Download and extract frames from videos on the internet
• Preview Video: Play extracted video frames in the ComfyUI interface
• Save Audio: Save audio data to file in various formats
• Save Video: Save video frames to file with optional audio

Installation

Manual Installation

1. Clone this repository into your ComfyUI custom_nodes directory:

cd ComfyUI/custom_nodes
git clone https://github.com/your-username/ComfyUI-MediaUtilities.git

2. Install the required dependencies:

cd ComfyUI-MediaUtilities
pip install -r requirements.txt

3. Restart ComfyUI

Usage

Load Audio From URL

This node allows you to load audio files directly from a URL.

Inputs:

• url: URL to the audio file (.mp3, .wav, .flac, .ogg, .m4a)

Outputs:

• AUDIO: Audio data that can be used with other ComfyUI audio nodes

Preview Audio

This node creates an audio player in the UI to preview audio data.

Inputs:

• audio: Audio data from an audio loader node

Load Video From URL

This node allows you to load video files directly from a URL and extract frames.

Inputs:

• url: URL to the video file (.mp4, .webm, .mkv, .mov, .avi)
• force_rate: Force a specific frame rate (0 to use original)
• force_size: Resize video to specific dimensions
• custom_width: Width for custom resizing
• custom_height: Height for custom resizing
• frame_load_cap: Maximum number of frames to load
• skip_first_frames: Number of frames to skip from the beginning
• select_every_nth: Select every nth frame

Outputs:

• FRAMES: Extracted video frames
• frame_count: Number of extracted frames
• audio: Audio track extracted from the video
• video_info: Video metadata (fps, duration, etc.)

Preview Video

This node creates a video player in the UI to preview extracted frames.

Inputs:

• frames: Frames from a video loader node
• fps: Playback frame rate
• video_info: Video metadata

Save Audio

This node saves audio data to a file in various formats.

Inputs:

• audio: Audio data to save
• filename_prefix: Base name for the output file
• format: Output format (mp3, wav, flac, ogg)
• sample_rate: Output sample rate
• normalize: Whether to normalize audio levels

Save Video

This node saves video frames to a file with optional audio.

Inputs:

• frames: Video frames to save
• fps: Frame rate for the output video
• filename_prefix: Base name for the output file
• format: Output format (mp4, webm, avi, mov, gif)
• codec: Video codec to use
• quality: Video quality (1-100)
• audio (optional): Audio to include in the video
• loop_count (optional): Number of times to loop the video
• pingpong (optional): Whether to play the video forward then backward

Outputs:

• video_path: Path to the saved video file

Example Workflows

Audio Loading and Preview

1. Add a "Load Audio From URL" node
2. Enter the URL of an audio file
3. Connect the output to a "Preview Audio" node
4. Execute the workflow to download and play the audio

Video Loading and Preview

1. Add a "Load Video From URL" node
2. Enter the URL of a video file
3. Configure the frame rate, size, and other options as needed
4. Connect the "FRAMES" output to a "Preview Video" node
5. Execute the workflow to download the video and preview the frames

Save Audio From URL

1. Add a "Load Audio From URL" node
2. Enter the URL of an audio file
3. Connect the output to a "Save Audio" node
4. Configure the output format and options
5. Execute the workflow to download and save the audio

Save Video From URL with Audio

1. Add a "Load Video From URL" node
2. Enter the URL of a video file
3. Connect the "FRAMES" output to a "Save Video" node
4. Connect the "audio" output to the "audio" input of the "Save Video" node
5. Configure the output format, codec, and other options
6. Execute the workflow to download and save the video with its audio

License

MIT

Credits

This project was inspired by:

• ComfyUI-VideoHelperSuite
• ComfyUI