ComfyUI-DeepFuze icon indicating copy to clipboard operation
ComfyUI-DeepFuze copied to clipboard

Hanging on: [DEEPFUZE.PROCESSORS.FRAME.MODULES.LIP_SYNCER] Download of the model is not done!

Open lord-lethris opened this issue 11 months ago • 1 comments

The node hangs on: [DEEPFUZE.PROCESSORS.FRAME.MODULES.LIP_SYNCER] Download of the model is not done!

D:\apps\SD-WebUI\ComfyUI\input D:\apps\SD-WebUI\ComfyUI\input\17376752259167092.wav
{'waveform': tensor([[[ 0.0000,  0.0000,  0.0000,  ..., -0.0011, -0.0014, -0.0020]]]), 'sample_rate': 22050}
gpen_bfr_512 30.0 video/h265-mp4
video h265-mp4
{'enhancer': 'gpen_bfr_512', 'frame_enhancer': 'None', 'face_mask_padding_left': 0, 'face_mask_padding_right': 0, 'face_mask_padding_bottom': 0, 'face_mask_padding_top': 0, 'trim_frame_start': 0, 'trim_frame_end': 0, 'device': 'cuda', 'frame_rate': 30.0, 'loop_count': 0, 'filename_prefix': 'deepfuzer', 'pingpong': False, 'save_output': False, 'images': ['27', 0], 'audio': ['30', 0], 'format': 'video/h265-mp4', 'pix_fmt': 'yuv420p10le', 'crf': 22, 'save_metadata': ['save_metadata', 'BOOLEAN', {'default': True}]}
D:\apps\SD-WebUI\ComfyUI\custom_nodes\ComfyUI-VideoHelperSuite\videohelpersuite\..\video_formats\h265-mp4.json
pix_fmt
crf
save_metadata
Output images were not of valid resolution and have had padding applied
D:\apps\SD-WebUI\ComfyUI\input\audio\17376752391866386.wav
['python', './run.py', '--frame-processors', 'lip_syncer', '-s', 'D:\\apps\\SD-WebUI\\ComfyUI\\input\\audio\\17376752391866386.wav', '-t', 'D:\\apps\\SD-WebUI\\ComfyUI\\temp\\deepfuzer_00001.mp4', '-o', 'D:\\apps\\SD-WebUI\\ComfyUI\\output\\deepfuze\\17376752391921763.mp4', '--trim-frame-start', '0', '--trim-frame-end', '832', '--face-mask-padding', '0', '0', '0', '0', '--headless', '--execution-providers', 'cuda']
[DEEPFUZE.PROCESSORS.FRAME.MODULES.LIP_SYNCER] Download of the model is not done!```

lord-lethris avatar Jan 23 '25 23:01 lord-lethris

Temporarily fixed it.

downloaded the file wav2lip_gan.onnx manually from: https://huggingface.co/bluefoxcreation/Wav2lip-Onnx/blob/main/wav2lip_gan.onnx

placed it in ../models/deepfuze

Opened ../custom-nodes/ComfyUI-DeepFuze/deepfuze/processors/frame/modules/lip_syncer.py And removed the following lines (at time of writing, this was lines 104 - 106)

if not deepfuze.globals.skip_download and not is_download_done(model_url, model_path):
		logger.error(wording.get('model_download_not_done') + wording.get('exclamation_mark'), NAME)
		return False

lord-lethris avatar Jan 24 '25 00:01 lord-lethris