ComfyUI Node
arkennemasis Narration Length (fit the shot to the voice)
Measure a narration clip and return the shot length that covers it, snapped to MiniMax H3's frame grid. Wire between the TTS node and the scene node so every shot outlasts its own voice-over.
arkennemasis Narration Length (fit the shot to the voice)
- narration
- length
- seconds
- report
- seconds_int
◄tail_seconds0.75►
◄max_seconds15.0►
◄lock_seconds0.0►
Categoryarkennemasis/Video
Inputs (4)
| Name | Type | Default | Description |
|---|---|---|---|
| narration | AUDIO | The rendered voice-over for THIS scene. | |
| tail_seconds | FLOAT | 0.750–5 | Picture held after the last word. Cutting on the syllable reads as a mistake; a beat reads as editing. |
| max_secondsopt | FLOAT | 15.05.2–15.1 | Hard ceiling. MiniMax H3 cannot exceed about 15.1 s in one shot, so a longer narration has to be shortened in the brief or split across two scenes. |
| lock_secondsopt | FLOAT | 0.00–15.1 | 0 = fit each shot to its own narration. Above 0, EVERY shot runs this long regardless of what is said — even pacing, at the cost of holding a frame under a short line. The value still snaps to H3's frame grid, so 10 s becomes 10.12 s. Keep the brief's word count under about 2 words per second of the locked length, or the voice will outrun the picture. |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| length | INT | — |
| seconds | FLOAT | — |
| report | STRING | — |
| seconds_int | INT | — |