Spaces:
Running
on
Zero
Running
on
Zero
Update README.md
Browse files
README.md
CHANGED
@@ -15,3 +15,115 @@ models:
|
|
15 |
- vrgamedevgirl84/Wan14BT2VFusioniX
|
16 |
- Kijai/WanVideo_comfy
|
17 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
15 |
- vrgamedevgirl84/Wan14BT2VFusioniX
|
16 |
- Kijai/WanVideo_comfy
|
17 |
---
|
18 |
+
## English Explanation
|
19 |
+
|
20 |
+
### Overview
|
21 |
+
This is a **VEO3 Free** application - an advanced AI video generation system that combines Wan2.1-T2V-14B model with automatic audio generation capabilities. It creates videos from text descriptions and automatically generates matching audio using MMAudio technology.
|
22 |
+
|
23 |
+
### Key Features
|
24 |
+
|
25 |
+
1. **Text-to-Video Generation**
|
26 |
+
- Uses Wan2.1-T2V-14B Diffusion model (14 billion parameters)
|
27 |
+
- Fast 4-step generation with NAG (Noise-Augmented Generation)
|
28 |
+
- Supports various resolutions from 128x128 to 896x896
|
29 |
+
- Duration: 1-8 seconds at 16 FPS
|
30 |
+
- Cinema-quality output with professional camera movements
|
31 |
+
|
32 |
+
2. **Automatic Audio Generation**
|
33 |
+
- MMAudio integration for synchronized sound effects
|
34 |
+
- Uses the same text prompt for both video and audio
|
35 |
+
- Configurable audio quality and guidance strength
|
36 |
+
- Optional feature - can be disabled if needed
|
37 |
+
|
38 |
+
3. **Advanced Controls**
|
39 |
+
- **NAG Scale**: Controls guidance strength (1.0-20.0)
|
40 |
+
- **Inference Steps**: Balances quality vs speed (1-8 steps)
|
41 |
+
- **Seed Control**: For reproducible results
|
42 |
+
- **Negative Prompts**: Specify what to avoid in generation
|
43 |
+
|
44 |
+
### How It Works
|
45 |
+
1. **Input**: Enter a detailed scene description
|
46 |
+
2. **Video Generation**: The AI creates video frames based on your prompt
|
47 |
+
3. **Audio Synthesis**: Automatically generates matching sound effects
|
48 |
+
4. **Output**: Combined video with synchronized audio
|
49 |
+
|
50 |
+
### Example Use Cases
|
51 |
+
- Film previews and concept visualization
|
52 |
+
- Music video creation
|
53 |
+
- Advertising content
|
54 |
+
- Creative storytelling
|
55 |
+
- Game cinematics
|
56 |
+
|
57 |
+
### Technical Details
|
58 |
+
- **GPU Acceleration**: Uses CUDA for fast processing
|
59 |
+
- **Model Architecture**: Transformer-based diffusion model
|
60 |
+
- **Audio Model**: Flow-matching based audio synthesis
|
61 |
+
- **Processing Time**: ~30-70 seconds depending on settings
|
62 |
+
|
63 |
+
### Tips for Best Results
|
64 |
+
- Use detailed, cinematic descriptions
|
65 |
+
- Include camera movements and visual style
|
66 |
+
- Specify lighting, colors, and atmosphere
|
67 |
+
- Add sound descriptions for better audio matching
|
68 |
+
- Higher NAG scale = more prompt adherence
|
69 |
+
|
70 |
+
---
|
71 |
+
|
72 |
+
## ํ๊ธ ์ค๋ช
|
73 |
+
|
74 |
+
### ๊ฐ์
|
75 |
+
**VEO3 Free**๋ Wan2.1-T2V-14B ๋ชจ๋ธ๊ณผ ์๋ ์ค๋์ค ์์ฑ ๊ธฐ๋ฅ์ ๊ฒฐํฉํ ๊ณ ๊ธ AI ๋น๋์ค ์์ฑ ์์คํ
์
๋๋ค. ํ
์คํธ ์ค๋ช
์ผ๋ก๋ถํฐ ๋น๋์ค๋ฅผ ์์ฑํ๊ณ MMAudio ๊ธฐ์ ์ ์ฌ์ฉํด ์๋์ผ๋ก ์ผ์นํ๋ ์ค๋์ค๋ฅผ ์์ฑํฉ๋๋ค.
|
76 |
+
|
77 |
+
### ์ฃผ์ ๊ธฐ๋ฅ
|
78 |
+
|
79 |
+
1. **ํ
์คํธ-๋น๋์ค ๋ณํ**
|
80 |
+
- Wan2.1-T2V-14B Diffusion ๋ชจ๋ธ ์ฌ์ฉ (140์ต ํ๋ผ๋ฏธํฐ)
|
81 |
+
- NAG(๋
ธ์ด์ฆ ์ฆ๊ฐ ์์ฑ)๋ฅผ ํตํ ๋น ๋ฅธ 4๋จ๊ณ ์์ฑ
|
82 |
+
- 128x128๋ถํฐ 896x896๊น์ง ๋ค์ํ ํด์๋ ์ง์
|
83 |
+
- ์ง์ ์๊ฐ: 16 FPS๋ก 1-8์ด
|
84 |
+
- ์ ๋ฌธ์ ์ธ ์นด๋ฉ๋ผ ์์ง์์ ํฌํจํ ์ํ ํ์ง ์ถ๋ ฅ
|
85 |
+
|
86 |
+
2. **์๋ ์ค๋์ค ์์ฑ**
|
87 |
+
- ๋๊ธฐํ๋ ์ฌ์ด๋ ํจ๊ณผ๋ฅผ ์ํ MMAudio ํตํฉ
|
88 |
+
- ๋น๋์ค์ ์ค๋์ค ๋ชจ๋ ๋์ผํ ํ
์คํธ ํ๋กฌํํธ ์ฌ์ฉ
|
89 |
+
- ์ค๋์ค ํ์ง๊ณผ ๊ฐ์ด๋์ค ๊ฐ๋ ์กฐ์ ๊ฐ๋ฅ
|
90 |
+
- ์ ํ์ ๊ธฐ๋ฅ - ํ์์ ๋นํ์ฑํ ๊ฐ๋ฅ
|
91 |
+
|
92 |
+
3. **๊ณ ๊ธ ์ ์ด ๊ธฐ๋ฅ**
|
93 |
+
- **NAG ์ค์ผ์ผ**: ๊ฐ์ด๋์ค ๊ฐ๋ ์ ์ด (1.0-20.0)
|
94 |
+
- **์ถ๋ก ๋จ๊ณ**: ํ์ง ๋ ์๋ ๊ท ํ ์กฐ์ (1-8๋จ๊ณ)
|
95 |
+
- **์๋ ์ ์ด**: ์ฌํ ๊ฐ๋ฅํ ๊ฒฐ๊ณผ๋ฅผ ์ํ ์ค์
|
96 |
+
- **๋ค๊ฑฐํฐ๋ธ ํ๋กฌํํธ**: ์์ฑ์์ ํผํ ์์ ์ง์
|
97 |
+
|
98 |
+
### ์๋ ๋ฐฉ์
|
99 |
+
1. **์
๋ ฅ**: ์์ธํ ์ฅ๋ฉด ์ค๋ช
์
๋ ฅ
|
100 |
+
2. **๋น๋์ค ์์ฑ**: AI๊ฐ ํ๋กฌํํธ ๊ธฐ๋ฐ ๋น๋์ค ํ๋ ์ ์์ฑ
|
101 |
+
3. **์ค๋์ค ํฉ์ฑ**: ์๋์ผ๋ก ์ผ์นํ๋ ์ฌ์ด๋ ํจ๊ณผ ์์ฑ
|
102 |
+
4. **์ถ๋ ฅ**: ๋๊ธฐํ๋ ์ค๋์ค๊ฐ ํฌํจ๋ ๋น๋์ค ์ถ๋ ฅ
|
103 |
+
|
104 |
+
### ํ์ฉ ์ฌ๋ก
|
105 |
+
- ์ํ ํ๋ฆฌ๋ทฐ ๋ฐ ์ปจ์
์๊ฐํ
|
106 |
+
- ๋ฎค์ง ๋น๋์ค ์ ์
|
107 |
+
- ๊ด๊ณ ์ฝํ
์ธ ์์ฑ
|
108 |
+
- ์ฐฝ์์ ์คํ ๋ฆฌํ
๋ง
|
109 |
+
- ๊ฒ์ ์๋ค๋งํฑ
|
110 |
+
|
111 |
+
### ๊ธฐ์ ์ฌ์
|
112 |
+
- **GPU ๊ฐ์**: ๋น ๋ฅธ ์ฒ๋ฆฌ๋ฅผ ์ํ CUDA ์ฌ์ฉ
|
113 |
+
- **๋ชจ๋ธ ์ํคํ
์ฒ**: ํธ๋์คํฌ๋จธ ๊ธฐ๋ฐ ํ์ฐ ๋ชจ๋ธ
|
114 |
+
- **์ค๋์ค ๋ชจ๋ธ**: ํ๋ก์ฐ ๋งค์นญ ๊ธฐ๋ฐ ์ค๋์ค ํฉ์ฑ
|
115 |
+
- **์ฒ๋ฆฌ ์๊ฐ**: ์ค์ ์ ๋ฐ๋ผ ์ฝ 30-70์ด
|
116 |
+
|
117 |
+
### ์ต์์ ๊ฒฐ๊ณผ๋ฅผ ์ํ ํ
|
118 |
+
- ์์ธํ๊ณ ์ํ์ ์ธ ์ค๋ช
์ฌ์ฉ
|
119 |
+
- ์นด๋ฉ๋ผ ์์ง์๊ณผ ์๊ฐ์ ์คํ์ผ ํฌํจ
|
120 |
+
- ์กฐ๋ช
, ์์, ๋ถ์๊ธฐ ๋ช
์
|
121 |
+
- ๋ ๋์ ์ค๋์ค ๋งค์นญ์ ์ํด ์ฌ์ด๋ ์ค๋ช
์ถ๊ฐ
|
122 |
+
- ๋์ NAG ์ค์ผ์ผ = ํ๋กฌํํธ์ ๋ ์ถฉ์คํ ์์ฑ
|
123 |
+
|
124 |
+
### ํน๋ณ ๊ธฐ๋ฅ
|
125 |
+
- **์ํ๊ธ ํ๋กฌํํธ ์์ **: ์ ๋ฌธ์ ์ธ ์ดฌ์ ๊ธฐ๋ฒ์ด ํฌํจ๋ 3๊ฐ์ง ์์ ์ ๊ณต
|
126 |
+
- **์ค์๊ฐ ์งํ ํ์**: ์์ฑ ๊ณผ์ ์ ์ค์๊ฐ์ผ๋ก ํ์ธ
|
127 |
+
- **์ํด๋ฆญ ์์ ์ ์ฉ**: ์์ ๋ฅผ ํด๋ฆญํ๋ฉด ์๋์ผ๋ก ์ค์ ๊ฐ ์ ์ฉ
|
128 |
+
|
129 |
+
์ด ๋๊ตฌ๋ ์ ๋ฌธ๊ฐ ์์ค์ ๋น๋์ค ์ฝํ
์ธ ๋ฅผ ์ฝ๊ฒ ์์ฑํ ์ ์๋๋ก ์ค๊ณ๋์์ผ๋ฉฐ, ์ฐฝ์์ ์ธ ์์ด๋์ด๋ฅผ ๋น ๋ฅด๊ฒ ์๊ฐํํ๋ ๋ฐ ์ด์์ ์
๋๋ค.
|