I posted this elsewhere so I'll put it here too: DINet - incredible in extremely limited contexts. Subject cannot be close-up, subject cannot turn even slightly from the camera. SadTalker video - can do close-ups but not middle or far distance, probably best if paired with a second pass through a face restorer. SadTalker image - same deal. VideoRetalker - I can't install this. FaceFusion - a really great package, works best for middle distance and not close-ups, does a lot of the steps together in one tool, doesn't include looping video for longer audio or video render cutoff at audio file end... Wav2Lip - great, video will loop if audio is longer, but output is lowest quality with oldest tech. TLDR: There is no single perfect option.
Yes, Jennifer. This is exactly where Lipsynthesis is moving! You will able to make video using our platform with your face and your voice to speak multiple languages.
@@khalil2099 That's true but I can still do upscale with other workflow the right mouth shape is still more important.. but well.. today I'm prefer RunwayML
best way to test it would be to use the same video for all 3 test.
This was just an ad
I posted this elsewhere so I'll put it here too:
DINet - incredible in extremely limited contexts. Subject cannot be close-up, subject cannot turn even slightly from the camera.
SadTalker video - can do close-ups but not middle or far distance, probably best if paired with a second pass through a face restorer.
SadTalker image - same deal.
VideoRetalker - I can't install this.
FaceFusion - a really great package, works best for middle distance and not close-ups, does a lot of the steps together in one tool, doesn't include looping video for longer audio or video render cutoff at audio file end...
Wav2Lip - great, video will loop if audio is longer, but output is lowest quality with oldest tech.
TLDR: There is no single perfect option.
What about geneface++ ?
How do I get FaceFusion
I'd argue that Retalking is far superior to your expensive software.
Hidden under a paywall great 👍
Can you upload audio instead of text to lipSynthesis ?
lipsysnthesis is expensive and you can only use there models unless you pay
Can we upload custom audio and video ?
where is he colab?
One thing I can say about this AI, practicing these prices won't get you very far
consigo fazer upload de um video meu e sincronizar voz com audio?
Yes, Jennifer. This is exactly where Lipsynthesis is moving! You will able to make video using our platform with your face and your voice to speak multiple languages.
Hidden under a paywall?????????????????? Where is the free trial?
Unlimited Access For $10 = fake 😠
What is fake? $10 monthly fee allows user to access any model in our database without no extra payment.
There are free softwares as well, so why $10 would be fake?!
@@AzizimIstanbulwhat are those free software?
you cant put your own model without asking? LOL
Wav2Lip offer more than blurry mouth, you can enhance them, and there also face fusion, your video r misleading
yeah but wav2lip teeth is not great. They have low res model.
@@khalil2099 That's true but I can still do upscale with other workflow the right mouth shape is still more important.. but well.. today I'm prefer RunwayML
@@vischutanan24 can runwayML do consistent character ? same face but different scene ? if yes, then it is great