# Prophetero Take Mixer GLOBAL — Picture from One Take, Sound from Another
Choose the picture from one video and the audio from another, adjust their independent starting positions, check a short preview and export an MP4 through ComfyUI.
Not limited to H3: supported clips can come from H3, LTX, other generators, cameras or editors. No generation model, LoRA, cloud account or GPU inference is required. Media assembly runs locally using FFmpeg.
## What you can do
- Choose or drop the picture and sound source videos.
- Set independent picture/audio start positions and the output duration.
- Preview a short section before the full export.
- Keep the source frame dimensions, crop to 1920 x 1080 or fit the full picture into that frame with bars where needed.
- Adjust the H.264 CRF quality setting.
Keeping source dimensions does not mean lossless picture copying. Take Mixer exports a newly encoded H.264/AAC MP4. It does not use Audio Fix's unchanged-picture export method. Your originals are not modified.
## Included and required
Includes the Prophetero workflow, custom node/interface, Windows installer, licence and guides. Requires an existing working ComfyUI/Python environment. The installer checks aiohttp and imageio-ffmpeg and asks before installing a missing dependency. Third-party software has its own terms.
1. Extract the entire ZIP and stop ComfyUI; stop SwarmUI too if it launches that backend.
2. Run START_HERE.cmd and select the ComfyUI folder containing main.py that your application actually uses.
3. Restart the backend and load Prophetero_Take_Mixer_GLOBAL.json.
4. Choose two short clips, set source positions and preview before exporting.
Outputs go under output/Prophetero/TakeMixer/Final in that ComfyUI installation. Uploaded files create local copies under output/Prophetero/TakeMixer/Inputs and consume disk space.
## Limits of this version — 1.0.1
The sound source must contain audio, and the selected sections must fit within both sources. This version supports SDR; HDR sources are rejected. The final duration control covers 0.25-120 seconds. Additional audio tracks, subtitles, chapters and source generation metadata are not preserved by this packaged version. Preview access is limited to uploaded files and files in ComfyUI input/output; other absolute paths can export but may not preview. Browser uploads are limited to 8 GiB.
Free personal and commercial creative use is permitted under the included Prophetero licence. Free sharing/modification is allowed under its terms; resale of Prophetero software or paid modified copies requires separate permission.
## Credits and original sources
- Media processing engine: [FFmpeg and its contributors](https://ffmpeg.org/).
- Python FFmpeg access: [imageio-ffmpeg and its contributors](https://github.com/imageio/imageio-ffmpeg).
- Local web integration: [aiohttp and its contributors](https://github.com/aio-libs/aiohttp).
- Host application: [ComfyUI and its contributors](https://github.com/Comfy-Org/ComfyUI).
## Scope of the Prophetero contribution
Prophetero provides the workflow integration, controls, installer and user interface/monitor supplied in this package. The underlying media engines and models are third-party projects, credited below. No claim is made that Prophetero trained those models or authored the upstream engines. No official endorsement is implied.
## Rights and dependencies
The included Prophetero licence applies only to Prophetero-authored material. Third-party code, models, libraries and FFmpeg builds retain their own terms; Prophetero's software resale restriction does not replace or restrict their original licences. See THIRD_PARTY.md and any notices retained inside bundled components. You must have the rights needed for your source media.