Alibaba EMO AI Model: The artificial intelligence sector is becoming more and more competitive. Alibaba, a Chinese business, has unveiled the EMO video AI model as a rival to Open AI’s Sora model. The latest Photo-to-Video AI model was unveiled by Alibaba’s Institute for Intelligent Computing Research and is proficient at producing audio-driven portrait movies.
OpenAI’s Sora and the Alibaba’s EMO video model largely offer similar functions and are comparable. Emote Portrait Alive is a program that uses an image and audio file to produce a short film. With this AI tool, users can make movies with a maximum duration of one minute and thirty seconds, during which the portrait can move, sing, and speak.
What’s Unique in Alibaba’s EMO AI Model
Alibaba’s EMO can make some of the famous painting such as Leonardo da Vinci and Monalisa speak, laugh and talk and even sing a song. This functionality will offer a unique way of entertainment to the users as they will be able to transform still photos into a moving video.
How Alibaba EMO AI Model Works?
The ability of EMO to alter a photo’s subject’s facial expressions is one of its finest features. In addition to this, subject’s lips can also be matched with original sounds which gives the impression that the film is real and first hand. Users will be also able to experiment with pictures, drawings, cartoons in the anime style with any still image.
How EMO Model Is Different From OpenAI Sora
| Feature | Alibaba EMO AI | OpenAI Sora |
|---|---|---|
| Function | Animates portraits and images | Generates videos from text descriptions |
| Technology | Audio-to-video synthesis (Diffusion Model) | Text-to-video generation (Diffusion Model) |
| Input | Single portrait image and audio clip | Text description |
| Output | Video of the portrait character speaking or singing with realistic facial expressions and lip sync | Video scene based on the text description |
| Strengths | Highly expressive and natural-looking facial animation, precise lip syncing | High-quality video generation, ability to create diverse scenes and landscapes |
| Weaknesses | Limited to animating existing portraits, requires audio input | Limited ability to generate realistic human characters, may not be as accurate in capturing details from the text description |
| Current Stage | Research prototype, not yet commercially available | Research project, not yet commercially available |
OpenAI’s Sora model reads the text and prepares the full movie. It can use a text prompt to produce an HD video. OpenAI Sora currently is accessible to limited number of people and not everyone can access it. It has been made available by the OpenAI to a small group of researchers. On the other hand, Alibaba EMO AI Model will be accessible to a wider audience. The EMO Model will offer a free and paid plans, while there is no free plan to use OpenAI Sora and it is only accessible premium users. This factor can prove an advantage for Alibaba EMO AI Model which can even make it a first choice of users in the near future.
Keep watching our YouTube Channel ‘DNP INDIA’. Also, please subscribe and follow us on FACEBOOK, INSTAGRAM, and TWITTER.