Video Merge Node

This custom ComfyUI node now uses native Comfy-style inputs:

• video_frames: IMAGE
• voice_audio: AUDIO
• background_audio: AUDIO
• srt_text: multiline STRING

The rendered video is saved into ComfyUI's default output directory.

What it does

1. takes an IMAGE batch as the video frame sequence
2. burns the provided SRT text into the frames
3. pads short audio with silence
4. trims long audio to the exact video duration
5. mixes voice and BGM into one track
6. exports a final .mp4 without extending the video tail, so no black screen is introduced

Built-in subtitle presets

• classic_white
• short_video_yellow
• cinematic_box
• news_blue
• minimal_clean
• karaoke_glow

Requirements

• ffmpeg

ffmpeg must be available in PATH, and it needs subtitle filter support.

Important note

Because IMAGE input is a frame batch instead of a video file, this node adds an fps parameter to define the output video speed and total duration.