ComfyUI Node
Smart Video Chunker
Silence-based video/audio chunking for LTX 2.3 / Wan 2.2. Looks for silence in audio to create chunk breaks. This is helpful when speakers take a breath and we don't cut off speaker while talking. Chunking is required to get around VRAM issues. For Low VRAM set chunk size lower
Smart Video Chunker
- video
- audio
- num_chunks
- chunkOImages
- chunkOfAudio
- numberFrames
- actual_end_time
- actual_start_time
- chunk_duration
- numGenerationFrames
◄index0►
◄chunk_secs10►
◄variation2►
◄source_fps30.00►
◄target_fps25.00►
◄start_time_override-1.000►
◄model_typeLTX►
CategoryTKNodes
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| video | IMAGE | Source video | |
| audio | AUDIO | Source audio | |
| index | INT | 00–9999 | Index from Loop - zero based |
| chunk_secs | INT | 10 | Size of each video segment in seconds |
| variation | INT | 2 | Num seconds variation. chunks_secs +/- variation adds flexiblity to find silence |
| source_fps | FLOAT | 30.001–240 | TRUE fps of incoming video tensor |
| target_fps | FLOAT | 25.001–240 | fps required by target model - 25 for LTX, 16 for WAN (typical) |
| start_time_overrideopt | FLOAT | -1.000-1–999999 | Use -1 for index 0, used to maintain exact timing of chunks. |
| model_typeopt | COMBO | LTX | Target model frame-count boundary: LTX=8n+1, WAN=4n+1 |
Outputs (8)
| Name | Type | Description |
|---|---|---|
| num_chunks | INT | Number Chunks Calculated for the Video. |
| chunkOImages | IMAGE | Video |
| chunkOfAudio | AUDIO | Audio |
| numberFrames | INT | # Frames after Snapping |
| actual_end_time | FLOAT | end time in video of chunk |
| actual_start_time | FLOAT | start time of chunk in video |
| chunk_duration | FLOAT | length of chunk |
| numGenerationFrames | INT | num Frames requested by user w/o snapping |