ComfyUI Node
🅛🅣🅧 Set Audio Video Mask By Time Split
Audio/video noise masks with separate time windows and temporal ramps (latent space). Differs from LTXVSetAudioVideoMaskByTime: no conditioning outputs; optional merge of prior video mask; spatial mask is blended with the envelope, not pasted.
🅛🅣🅧 Set Audio Video Mask By Time Split
- av_latent
- model
- vae
- audio_vae
- spatial_mask
- av_latent
◄video_fps24.00►
◄video_start_time0.00►
◄video_end_time10.00►
◄video_slope_len3►
◄mask_videotrue►
◄mask_init_value_video0.00►
◄audio_start_time0.00►
◄audio_end_time10.00►
◄audio_slope_len3►
◄mask_audiotrue►
◄mask_init_value_audio0.00►
◄merge_existing_video_maskfalse►
Categorylightricks/LTXV
Inputs (17)
| Name | Type | Default | Description |
|---|---|---|---|
| av_latent | LATENT | — | |
| model | MODEL | — | |
| vae | VAE | — | |
| audio_vae | VAE | — | |
| video_fps | FLOAT | 24.000–500 | — |
| video_start_time | FLOAT | 0.000–2000 | — |
| video_end_time | FLOAT | 10.000–2000 | — |
| video_slope_len | INT | 31–100 | — |
| mask_video | BOOLEAN | true | — |
| mask_init_value_video | FLOAT | 0.000–1 | — |
| audio_start_time | FLOAT | 0.000–2000 | — |
| audio_end_time | FLOAT | 10.000–2000 | — |
| audio_slope_len | INT | 31–100 | — |
| mask_audio | BOOLEAN | true | — |
| mask_init_value_audio | FLOAT | 0.000–1 | — |
| spatial_maskopt | MASK | Optional (T,H,W) or (H,W) mask; blended with temporal envelope for video (not a hard replace like LTXVSetAudioVideoMaskByTime). | |
| merge_existing_video_maskopt | BOOLEAN | false | If the latent already has noise_mask, multiply the new video mask by the existing per-frame scalar (same behavior as LTXVSetAudioVideoMaskByTime for video only). |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| av_latent | LATENT | — |