Thanks for all our awesome tutorials!
Thanks for all our awesome tutorials!
Fty
UHi. I'm getting this error trying to train voice with DL-Art-School:
Loading from ../experiments/dvae.pth
Traceback (most recent call last):
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\train.py", line 398, in <module>
trainer.init(args.opt, opt, args.launcher)
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\train.py", line 146, in init
self.model = ExtensibleTrainer(opt)
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\trainer\ExtensibleTrainer.py", line 105, in init
step = ConfigurableStep(step, self.env)
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\trainer\steps.py", line 49, in init
self.injectors.append(create_injector(injector, env))
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\trainer\inject.py", line 65, in create_injector
return injectors[opt_inject['type']](opt_inject, env)
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\trainer\injectors\audio_injectors.py", line 178, in init
self.dvae = load_model_from_config(cfg, dvae_name, device=f'cuda:{env["device"]}').eval()
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\utils\util.py", line 497, in load_model_from_config
sd = torch.load(load_path, map_location=device)
File "C:\Users\hasan\miniconda3\envs\DLAS\lib\site-packages\torch\serialization.py", line 777, in load
with _open_zipfile_reader(opened_file) as opened_zipfile:
File "C:\Users\hasan\miniconda3\envs\DLAS\lib\site-packages\torch\serialization.py", line 282, in init
super(_open_zipfile_reader, self).init(torch._C.PyTorchFileReader(name_or_buffer))
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory
Press any key to continue . . .
Loading from ../experiments/dvae.pth
Traceback (most recent call last):
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\train.py", line 398, in <module>
trainer.init(args.opt, opt, args.launcher)
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\train.py", line 146, in init
self.model = ExtensibleTrainer(opt)
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\trainer\ExtensibleTrainer.py", line 105, in init
step = ConfigurableStep(step, self.env)
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\trainer\steps.py", line 49, in init
self.injectors.append(create_injector(injector, env))
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\trainer\inject.py", line 65, in create_injector
return injectors[opt_inject['type']](opt_inject, env)
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\trainer\injectors\audio_injectors.py", line 178, in init
self.dvae = load_model_from_config(cfg, dvae_name, device=f'cuda:{env["device"]}').eval()
File "C:\Users\hasan\voice-clone\DL-Art-School\codes\utils\util.py", line 497, in load_model_from_config
sd = torch.load(load_path, map_location=device)
File "C:\Users\hasan\miniconda3\envs\DLAS\lib\site-packages\torch\serialization.py", line 777, in load
with _open_zipfile_reader(opened_file) as opened_zipfile:
File "C:\Users\hasan\miniconda3\envs\DLAS\lib\site-packages\torch\serialization.py", line 282, in init
super(_open_zipfile_reader, self).init(torch._C.PyTorchFileReader(name_or_buffer))
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory
Press any key to continue . . .
Fwhat is your gpu
Falso
Funinstall all python
Finstall python 3.10.x into C drive
Finstall voice clan into C drive
Fnot in users folder
UI think there was some files that are not downloaded correctly. When I delete everything and install again, it worked.
WSir, does VGA/GPU affect the quality of deepfake?
Fnope just speed
Ihas anybody done video with gpt4v? can we? extract image, combine RAG or something?
can we somehow use dall e to create video? create sequence of image and 'stictch' them together?
can we somehow use dall e to create video? create sequence of image and 'stictch' them together?
Fi doubt that you can do
Iwhy not? thanks for the response
Fit can generetate individual images
Fnot consisten to make video
Ii mean video2text is possible?
workflow is: extract image out of video, image2text using GPT4V, combining text together with prompt engineering, store them in embedding/vector db, finetune a model with those embeddings, then generate such images with DALL-E 3, consume 3rd party API to stitch them together?
workflow is: extract image out of video, image2text using GPT4V, combining text together with prompt engineering, store them in embedding/vector db, finetune a model with those embeddings, then generate such images with DALL-E 3, consume 3rd party API to stitch them together?
Ii think openAI has new feature to make DALL-3 more consistent now
Iat least a slide show now
Fyes possible
Feasy
Fwith GPT4 vision API
Ithanks, any feedback on my approach? cuz vision api is mostly for images not videos, ive not tried the other way round though, whether image2text (gpt4v) and text2image (dall-3) can be made consistent, with the new feature?
Fyou need to extract
Fvideo frames
Ithanks
Iwill try
Iany chance you do a video for this
Fi dont have
W
Ffor now royalty music is better
PFor sentimate analysis which one would you prefer ALBERT or DIstilBert? I want to run them in a relatively low resource env and want really good results.
Are there other new models that can are better? i don't want to call an API for this.
Are there other new models that can are better? i don't want to call an API for this.
Lhello
what faceswaper is the best?
and which one works best with pets?
also if i want to generate my kitty into Disney character which just one photo which controllent would u recommend ?
what faceswaper is the best?
and which one works best with pets?
also if i want to generate my kitty into Disney character which just one photo which controllent would u recommend ?
Flet me check
F,
message.txt2.04KB
Mi dont know if it is helpful for you guys there is a discord server that make ai aniamtions from text
i dont know i can use discord link so i send website link
https://www.mootion.com/landing
i dont know i can use discord link so i send website link
https://www.mootion.com/landing
Mits under develop
SCan't wait to fine tune this thing! https://stability.ai/news/stable-video-diffusion-open-ai-video-model
Stability AI
Stable Video Diffusion is a proud addition to our diverse range of open-source models. Spanning across modalities including image, language, audio, 3D, and code, our portfolio is a testament to Stability AI’s dedication to amplifying human intelligence.
Tdid you run it?
SNo not yet, have you?
Iwhats your most updated course? is it on udemy? or youtube? or patreon
Iquite confused
Icuz you have many channels
Iwhere can i find the google colab version
Fyoutube + patreon
Fyou need colab version of what?
Iim going through your udemy courses, probably need each one, because i dont have GPU on my local machine