Of course, I couldn't pass the opportunity to try out the new AI model Gemini 2.5 Flash Image edit better known as "nano-banana".
Here are couple of use cases. Replace character in an iconic meme.
@FAL
@deifosv@synclabs Did you try some other models for lipsync? IMO Pixverse Lipsync and Tavus Hummingbird Lipsync v0 provide better results while also being cheaper.
@FAL I definitely want to give credit to @MediaCentralDev
His post https://t.co/jJgOwQyHgI inspired me to do my own research and provide exact information about the pricing of the AI models and the actual cost of the generated content.
I tested all of the Best Avatar Models on @FAL to find which creates the most realistic product in hand videos.
Setup:
• Generated a product shot (lady + body lotion) using Wan 2.2 Instagirl LoRA, Ideogram Character and Qwen Image.
• Created a custom voice with MiniMax Speech-02 HD
• Generated a silent video with Wan 2.2 image-to-video
Which wins? Let's see 🧵👇
What are the best AI models that you can use to generate video with lipsync while having total control over the audio?
I decided to use @FAL platform to test the best models in one place.
@FAL Why didn't I include the Veo3 model? Simply because you don't have much control over what the audio output will be.
The pricing is quite steep. For every second it costs $0.50 (audio off) or $0.75 (audio on) -> 10s video with audio will cost $7.5
@FAL VEED/lipsync
Input: video + audio
Pricing: $0.4 per minute of the generated video
Video cost: $0.40
Video quality: 0/10 - audio totally out of sync
It took 68s to generate the video.