# Prophetero Audio Fix — Better Sound. Original Picture.
You finally got the picture right. Now adjust the sound without re-encoding the final video picture.
## Not just for H3 videos
Use supported local videos from LTX, H3, other AI generators, cameras or editors. Compatibility depends on file format and streams, not the model that created the clip. A Civitai base-model label does not limit Audio Fix to H3.
## What you can do
- Increase or decrease volume.
- Apply two-pass loudness normalization with an adjustable target.
- Replace sound using an audio file or another video's audio.
- Move audio earlier or later, add audio fades, or mute the video.
- Export processed audio as WAV.
- Compare short original/processed previews before saving the full clip.
## Keep the picture you like
Full video exports copy the original encoded video stream and verify its encoded packet payload before publishing the output. Audio is processed separately. The output container and complete file are not byte-identical to the source, but the final video picture is not re-encoded. Originals stay untouched.
Short previews use compressed picture for speed; they are not the final export.
## Included and required
Includes the Prophetero workflow/node, drag-and-drop/file selection, Windows installer, English/Hungarian guides and controls/help in EN, HU, DE, FR, ES, IT, PT and PL. English is the default.
Requires a working ComfyUI backend and its Python environment. No generation model, cloud account or GPU generation is needed. FFmpeg and Python dependencies perform local media processing; missing dependencies are separate downloads, with installer approval before installation.
1. Extract the entire ZIP and run START_HERE.cmd.
2. Select the backend actually used by your application. For SwarmUI, use its ComfyUI backend.
3. Fully restart that backend after installation; refreshing the page alone does not load a newly installed custom node.
4. Open Prophetero_Audio_Fix_GLOBAL.json, choose a short clip and compare the previews. Run/Queue saves the full output.
Final outputs: output/Prophetero/AudioFix/Final under your ComfyUI installation.
## Limits — v0.1.0
Supports one primary video stream and one mono/stereo audio stream. Extra tracks, subtitles and data streams are not included. Auto chooses MP4 or MKV; a forced incompatible MP4 fails instead of re-encoding the picture. Metadata preservation depends on the container. Browser uploads create local copies and use disk space. See README_EN.txt for duration, upload and format limits.
Free personal/commercial creative use and free sharing/modification are allowed under the included licence. Reselling the Prophetero software, paid modified copies or paid bundles containing it requires separate permission. Source-media rights remain your responsibility.
## 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).
- Video stream inspection/verification: [PyAV and its contributors](https://github.com/PyAV-Org/PyAV).
## 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.
Description
Prophetero Audio Fix GLOBAL — Repair the sound. Keep the picture.
The picture is finally right, but a beep, a quiet passage or a distracting sound spoils it. Fix a selected passage instead of replacing the complete soundtrack, while keeping the original encoded picture in the final video.
NEW IN 0.3.0 — REPAIR PLANS
• A large repair workspace, opened from the existing ComfyUI node.
• Up to 50 named repairs, with independent on/off switches, editing, removal and undo.
• Replace an interval with an excerpt from the same original video or another audio/video file.
• Request 3, 5 or 8 same-video donor suggestions and listen to each separately before choosing.
• Local mute, attenuation and Gain (-36 to +24 dB). Each repair covers 0.02–30 seconds.
• Before/After at the same preview position, plus previews near the start/end of a repair.
• JSON project save/load and automatic local browser drafts, including unfinished editor values.
• All enabled repairs in one processing pass, followed by one final lossy audio encode.
THE ORIGINAL TOOLS REMAIN
Whole-track volume adjustment, two-pass loudness normalization, soundtrack replacement, timing offset, fades, mute and WAV extraction remain available in the node. They are separate from the interval-repair workspace.
NOT JUST FOR H3
Use supported local videos from LTX, H3, other generators, cameras or previous edits. Any H3 base-model label on this listing is a category, not a technical dependency. No generation model, cloud account or GPU generation is required.
KEEP THE PICTURE
Final video exports copy the original encoded video stream without picture re-encoding and verify its packet payload with SHA-256. Audio is processed separately and re-encoded as AAC, including the audio outside repaired intervals. Original files stay untouched; results get unique filenames. The short browser preview uses compressed picture for speed and is not the final picture export.
INSTALL OR UPDATE
1. Extract the complete ZIP and run START_HERE.cmd on Windows.
2. Select the ComfyUI backend your application actually uses. SwarmUI and portable parent folders are recognized when unambiguous; otherwise select the backend itself.
3. The installer asks before installing missing dependencies and backs up replaced files. No models are downloaded.
4. Fully restart ComfyUI, and SwarmUI if it starts that backend, then refresh the browser. A browser refresh alone does not load new server code.
5. Open Prophetero_Audio_Fix_GLOBAL.json. Click “Repair selected audio interval…” to open the new workspace.
6. Add/Update repairs in the list, make a processed preview, listen, then use the workspace's full-save button.
This ZIP is a complete package, suitable for a new installation or an update. Existing Audio Fix workflow field order remains compatible. The Windows installer is supplied; Linux/macOS installation is not validated in this release.
Final output: ComfyUI/output/Prophetero/AudioFix/Final
Eight interface/help languages: EN, HU, DE, FR, ES, IT, PT, PL. English default; English and Hungarian guides included.
PRACTICAL LIMITS
Suggestions match approximate level/texture, not the meaning of speech or musical rhythm. Search covers at most ten minutes near the fault. You choose and listen to the result; suggestions are not a promise of a natural repair. Enabled targets cannot overlap. Internal donors always read the original sound and cannot overlap an enabled repair. Local mute keeps the video timeline; it does not cut frames.
Previews last up to four seconds. Before/After can wait for media to load; it is not a gapless mixer. Full repair export requires applied edits and a current processed preview, but listening remains your responsibility. Repair-project JSON loads in the repair workspace, not the ComfyUI workflow loader; it contains paths/settings, not media.
One primary video and the first mono/stereo audio stream are supported, with a readable duration up to one hour and browser uploads up to 8 GiB. Extra audio tracks, subtitles and data streams are not carried over. Container/metadata compatibility still matters. Gain does not repair distortion or remove underlying noise.
TESTING
Automated synthetic tests cover combined beep replacement, mute, gain and external donors; one final audio encode; unchanged encoded picture, duration and metadata; overlap checks; simulated UI state; and installer behavior. The newest multi-repair controls still need a complete live-browser and real multi-fault listening check. Begin with a short test clip and share feedback.
CREDITS AND USE
Prophetero supplies the workflow, custom node and repair workspace. Media processing uses FFmpeg, with PyAV and imageio-ffmpeg; ComfyUI is installed separately. Upstream links and independent terms are in THIRD_PARTY.md. No model weights or third-party media are bundled.
Free personal and commercial creative use under the included Prophetero Audio Fix Free Use License. Free sharing/modification requires retaining that license. Resale, paid access or paid bundles of the software require separate written permission. These restrictions concern the software, not your permitted finished creative work. See LICENSE.txt.
If it saves a good take, let me know! A reaction, a short report or one practical suggestion helps shape the next update.
