#AI #CHATGPT #GEN-1
Since the beginning of the year, various AI services (Chat GPT, Stable Diffusion, etc.) that generate texts and pictures have been a constant topic of conversation.
Recently, technology that is truly the final version has emerged, and now it is possible to produce a video with AI alone.
Are there two main components of the video, voice and video?
So the first AI I will introduce is a voice AI.
It is a technique that learns the voice of a specific person and makes a voice very similar to the actual voice.
The degree of similarity varies depending on the amount and method of learning data, but when I used it myself, it was almost similar unless I decided to analyze it. (Note that we have used an AI service called Elevenlabs.)
As it is so similar, the number of cases of deep-fake voice phishing crimes using such voice AI in foreign countries has been increasing recently.
The second AI I'm going to introduce is a video AI that makes videos.
Runway's AI service, GEN-1, allows you to add references to your original video to create a whole new video.
As a reference, you can put images, presets, and texts, and if you find the right reference, you can make a movie with videos taken in the corner of the room.
Even the latest version of GEN-2, which was released by the developer, says that a new image can be created by simply entering text without the original image.
It was a shocking technology for creating video content.
The "Omokgyo Electronic Shopping Center" looked into voice AI and video AI, which are developing faster than expected.
0:00 Video Shooting AI Levels Today
0:48 President Biden celebrate?!
1:33 AI How to Make Voice Yourself
2:39 AI voice limited
4:58 AI made video
5:20 AI How to Make Your Own Video
6:52 AI video production tips!
8:30 Why we keep introducing AI
Негізгі бет [video AI, voiceAI] In the future, video production time will be reduced by 100 times.
Пікірлер: 171