Recently, AI to do video wind up, the emergence of a lot of excellent software, some of the effects have reached the commercial level, today and we will talk about theVideo AI softwareSee later for a list of recommended software.
one,AI Video General Knowledge
What is AI doing video?
Video incorporates text, voice, images and other multi-dimensional content
"AI do video" refers to the use of AI technology to complete all aspects of video creation, including: topic selection, script creation, split-screen design, video generation, different angles of filming, special effects, dubbing and music.
Stages of Development for Generative AI Video
Current mainstream video AI is generative AI
After many upgrades and iterations of text-to-writing and text-to-graphics, we are at a critical historical point of gradual penetration of AI+productivity office and design, AI+video, and AI+3D
The ways AI can do video include text-born video, graph-born video, and video-born video
Common AI video software
Video is a comprehensive result of topic selection, script creation, split-screen design, video generation, filming from different angles, special effects, voice-overs and soundtracks.
(i) Track selection
You can choose to cash in on video numbers or sand animation, and later we'll cover newscasts, commercials, and film production
(ii) Selection of themes
Pick a topic you're familiar with
(iii) Script writing
Scripts are primarily created using AI macromodels and need to incorporate cue words
Text AI can help define a theme, write a drawing or make a cue for a video
DeepSeek and Beanbag have a better "net feel".
- DeepSeek: chat.deepseek.com
- Ask White (spare tire): www.wenxiaobai.com
- Byte Jumping Beanbag: www.doubao.com
(1) Beanbag can automatically extract the text of a single Shakeology video based on the link to that video
(2) Beanbags can read pictures and write different shots of the same character based on them
You can also let the AI read the pictures and generate cues, Kimi, Qianqi read the pictures (deepseek can't analyze the pictures)
(iv) Video analysis
geimin 1.5: the only good video analyzer that can analyze video cue words frame by frame aistudio.google.com
Students in China can use the new Qianqian 3.0 chat.qwen.ai
Frame grabber: Ezgif ezgif.com/video-to-jpg
(v) Image generation
Graph-born video is by far the most manageable way to make a graph and then a video with it
1. National production
Domestically produced i.e. Dream AI, Keling, Tongyi Wanxiang
Preferred that dream, free points more, fast, the new 3.0 support picture Chinese (50 words)
- Byte jumping namely dream: jimeng.jianying.com
- Doubao: doubao.com Doubao integration is also namely dream, but can be a cue out of a dozen diagrams
- Racer Koring: klingai.kuaishou.com
- Tongyi Wanxiang: ongyi.aliyun.com/wanxiang
- Rhinoceros: www.eqmj.cc
2. Overseas
You can also draw split-screen designs, Imagen3 labs.google
- Imagen3 is supposed to be the best AI in the world, and you need to apply for Google Mail to complete the registration.
- Flux: glif.app, free generation of fux pro1.1 ultra
- krea: krea.ai
- chatgpt4o Words are not enough chatgpt.com
- Ideogram AI: ideogram.ai, which generates IP characters
(vi) Video AI
The world's first choice that dream, free points more, fast, the new 3.0 support cue word modification sub-scene. That dream represents the highest level in the world!
Kerin is the most versatile; conch, tongyi, vidu, step, beanbag
Domestic
1, byte jumping that dream: jimeng.jianying.com
2. Racer Ke Ling: klingai.kuaishou.com
3. Conch: hailuoai.com
4、Tongyi Wanxiang (video with sound effects): tongyi.aliyun.com/wanxiang
5. Vidu (can't do graphs, can do videos): www.vidu.cn
6, Aishi Technology's (old school video AI) PixVerse: app.pixverse.ai
7. Smart Spectrum Clear Shadow: chatglm.cn/video
Overseas
1. OpenAI's Sora: sora.com
2, google veo 2:
aistudio.google.com/app/generate-video
veo2 works very well and can be used for free, but you need to apply for Google mail, click here to see the application tutorials
3. Luma :lumalabs.ai/dream-machine
4. Runway's Gen-3:app.runwayml.com
5. Pika: pika.art
6. Stable Video Diffusion: stability.ai/stable-video
(vii) Speech generation
1, Conch voice (completely free; support 1 time 6000 words): hailuoai.com/audio
2, fish tts (can clone the voice): fish.audio
3. RuiSheng Technology website (temporarily closed): www.reecho.cn
4. Elevenlabs: try.elevenlabs.io
5. chattts: chattts.com (unofficial)
(viii) Sound effects
1、Tongyi千问 tongyi.aliyun.com/wanxiang：Automatic dubbing according to video content
2. vidu2: www.vidu.studio Sound effects can be generated manually on a second-by-second basis
3. Uncopyrighted sound effects website: pixabay.com, you need to manually search by yourself
(ix) Counterparts
1, that is, the dream, can Ling, have the lip-synching function, but currently only supports 5-10 seconds
2. dreamface, the free version can only be 30 seconds at a time, 3 minutes after payment.
(x) Soundtrack
1. i.e. dream: jimeng.jianying.com
2. Skyworks Mureka: www.mureka.ai
3. Suno
(xi) Other support
Photo HD: krea.com
Keying: remove.bg or dream canvas
(xii) Digital people
dreamface, Zongshang, Cicada Mirror, Xunfei Zhisuo, Jiechuang, hengen, etc.
(xiii) Editing and compositing
Cutting and Screening, Wanxing Meow Shadow, Zhi Shadow Online Editing zenvideo.qq.com
It's time to learn AI for video!
If you are related to the film and television industry, you can use AI to expand your professional edge; if you are a professional, you can use AI to quickly complete the short video production, assisting in business development and daily publicity; if you are a short video blogger, you can utilize AI to create creative videos, attracting streams and realizing cash.
You can see that the AI video space is changing to a major new technology almost every 1-2 months because the technology is advancing, the functionality is improving, and it's not just about the technology, it's about the functionality.But it's also this technology iteration that creates the bonus periodThe first thing you need to do is to get your own share of the pie, just like Jitterbug in 2021, as long as you get your feet wet and don't get down on the table.
Get in three months early and there aren't as many good tools; get in three months late and you may miss the best opportunities
Take action to seize the wind mouth of "generative AI + video"! Seize these months, practicing AI+Video skills, finding business models, and realizing fast cash are the most important choices now, so that we can be one step ahead of others when more and better software appears later!