Researchers have introduced MoCha, a novel AI model, developed with the participation of Meta* AI specialists. This system is designed to create cinematic-quality videos of people using both text descriptions and audio inputs.
The system requires an audio file to operate. Based on this audio, the AI synchronizes the articulation of virtual characters with speech. MoCha is also capable of generating full dialogue scenes with multiple characters. However, at the time of publication, the technology has not yet reached full cinematic standards.
Finley Holt