新增示例库逻辑
This commit is contained in:
puke
57
README.md
57
README.md
@@ -34,7 +34,7 @@
|
||||
- ✅ **全自动生成** - 输入主题，3 分钟自动生成完整视频
|
||||
- ✅ **AI 智能文案** - 根据主题智能创作解说词，无需自己写脚本
|
||||
- ✅ **AI 生成配图** - 每句话都配上精美的 AI 插图
|
||||
- ✅ **真人语音** - 100+ 种真人声音可选，告别机械音
|
||||
- ✅ **AI 生成语音** - 支持 Edge-TTS、Index-TTS 等众多主流 TTS 方案
|
||||
- ✅ **背景音乐** - 支持添加 BGM，让视频更有氛围
|
||||
- ✅ **视觉风格** - 多种模板可选，打造独特视频风格
|
||||
- ✅ **灵活尺寸** - 支持竖屏、横屏等多种视频尺寸
|
||||
@@ -123,16 +123,7 @@ uv run streamlit run web/app.py
|
||||
- **固定文案内容**: 直接输入完整文案，跳过 AI 创作
|
||||
- 适合：已有现成文案，直接生成视频
|
||||
|
||||
---
|
||||
|
||||
### 🎵 音频设置（左侧栏）
|
||||
|
||||
#### 语音选择
|
||||
- 从下拉菜单选择解说声音
|
||||
- 提供 4 种精选声音（男声/女声、专业/年轻）
|
||||
- 点击「试听语音」可以预览效果
|
||||
|
||||
#### 背景音乐
|
||||
#### 背景音乐（BGM）
|
||||
- **无 BGM**: 纯人声解说
|
||||
- **内置音乐**: 选择预置的背景音乐（如 default.mp3）
|
||||
- **自定义音乐**: 将你的音乐文件（MP3/WAV 等）放到 `bgm/` 文件夹
|
||||
@@ -140,16 +131,40 @@ uv run streamlit run web/app.py
|
||||
|
||||
---
|
||||
|
||||
### 🎤 语音设置（中间栏）
|
||||
|
||||
#### TTS 工作流
|
||||
- 从下拉菜单选择 TTS 工作流（支持 Edge-TTS、Index-TTS 等）
|
||||
- 系统会自动扫描 `workflows/` 文件夹中的 TTS 工作流
|
||||
- 如果懂 ComfyUI，可以自定义 TTS 工作流
|
||||
|
||||
#### 参考音频（可选）
|
||||
- 上传参考音频文件用于声音克隆（支持 MP3/WAV/FLAC 等格式）
|
||||
- 适用于支持声音克隆的 TTS 工作流（如 Index-TTS）
|
||||
- 上传后可以直接试听
|
||||
|
||||
#### 预览功能
|
||||
- 输入测试文本，点击「预览语音」即可试听效果
|
||||
- 支持使用参考音频进行预览
|
||||
|
||||
---
|
||||
|
||||
### 🎨 视觉设置（中间栏）
|
||||
|
||||
#### 视觉风格
|
||||
#### 图像生成
|
||||
决定 AI 生成什么风格的配图。
|
||||
|
||||
**ComfyUI 工作流**
|
||||
- 选择图像生成的工作流文件
|
||||
- 从下拉菜单选择图像生成工作流
|
||||
- 支持本地部署（selfhost）和云端（RunningHub）工作流
|
||||
- 默认使用 `image_flux.json`
|
||||
- 如果懂 ComfyUI，可以放自己的工作流到 `workflows/` 文件夹
|
||||
|
||||
**图像尺寸**
|
||||
- 设置生成图像的宽度和高度（单位：像素）
|
||||
- 默认 1024x1024，可根据需要调整
|
||||
- 注意：不同的模型对尺寸有不同的限制
|
||||
|
||||
**提示词前缀（Prompt Prefix）**
|
||||
- 控制图像的整体风格（语言需要是英文的）
|
||||
- 例如：Minimalist black-and-white matchstick figure style illustration, clean lines, simple sketch style
|
||||
@@ -158,7 +173,7 @@ uv run streamlit run web/app.py
|
||||
#### 视频模板
|
||||
决定视频画面的布局和设计。
|
||||
|
||||
- 从下拉菜单选择模板（default.html、modern.html、classic.html 等）
|
||||
- 从下拉菜单选择模板，按尺寸分组显示（竖屏/横屏/方形）
|
||||
- 点击「预览模板」可以自定义参数测试效果
|
||||
- 如果懂 HTML，可以在 `templates/` 文件夹创建自己的模板
|
||||
|
||||
@@ -190,9 +205,9 @@ A: 生成一个 3 段视频大约需要 2-5 分钟，取决于你的网络和 AI
|
||||
**Q: 视频效果不满意怎么办？**
|
||||
A: 可以尝试：
|
||||
1. 更换 LLM 模型（不同模型文案风格不同）
|
||||
2. 调整提示词前缀（改变配图风格）
|
||||
3. 更换语音（不同声音适合不同内容）
|
||||
4. 尝试不同的视频模板
|
||||
2. 调整图像尺寸和提示词前缀（改变配图风格）
|
||||
3. 更换 TTS 工作流或上传参考音频（改变语音效果）
|
||||
4. 尝试不同的视频模板和尺寸
|
||||
|
||||
**Q: 费用大概多少？**
|
||||
A: **本项目完全支持免费运行！**
|
||||
@@ -237,11 +252,3 @@ Pixelle-Video 的设计受到以下优秀开源项目的启发：
|
||||
|
||||
[](https://star-history.com/#PixelleLab/Pixelle-Video&Date)
|
||||
|
||||
---
|
||||
|
||||
<div align="center">
|
||||
<p>Made with ❤️ by PixelleLab</p>
|
||||
<p>
|
||||
<a href="#top">回到顶部 ⬆️</a>
|
||||
</p>
|
||||
</div>
|
||||
|
||||
@@ -1,343 +0,0 @@
|
||||
# Pixelle-Video Capabilities Guide
|
||||
|
||||
> Complete guide to using LLM, TTS, and Image generation capabilities
|
||||
|
||||
## Overview
|
||||
|
||||
Pixelle-Video provides three core AI capabilities:
|
||||
- **LLM**: Text generation using LiteLLM (supports 100+ models)
|
||||
- **TTS**: Text-to-speech using Edge TTS (free, 400+ voices)
|
||||
- **Image**: Image generation using ComfyKit (local or cloud)
|
||||
|
||||
## Quick Start
|
||||
|
||||
```python
|
||||
from pixelle_video.service import pixelle_video
|
||||
|
||||
# LLM - Generate text
|
||||
answer = await pixelle_video.llm("Summarize 'Atomic Habits' in 3 sentences")
|
||||
|
||||
# TTS - Generate speech
|
||||
audio_path = await pixelle_video.tts("Hello, world!")
|
||||
|
||||
# Image - Generate images
|
||||
image_url = await pixelle_video.image(
|
||||
workflow="workflows/book_cover_simple.json",
|
||||
prompt="minimalist book cover design"
|
||||
)
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 1. LLM (Large Language Model)
|
||||
|
||||
### Configuration
|
||||
|
||||
Edit `config.yaml`:
|
||||
|
||||
```yaml
|
||||
llm:
|
||||
default: qwen # Choose: qwen, openai, deepseek, ollama
|
||||
|
||||
qwen:
|
||||
api_key: "your-dashscope-api-key"
|
||||
base_url: "https://dashscope.aliyuncs.com/compatible-mode/v1"
|
||||
model: "openai/qwen-max"
|
||||
|
||||
openai:
|
||||
api_key: "your-openai-api-key"
|
||||
model: "gpt-4"
|
||||
|
||||
deepseek:
|
||||
api_key: "your-deepseek-api-key"
|
||||
base_url: "https://api.deepseek.com"
|
||||
model: "openai/deepseek-chat"
|
||||
|
||||
ollama:
|
||||
base_url: "http://localhost:11434"
|
||||
model: "ollama/llama3.2"
|
||||
```
|
||||
|
||||
### Usage
|
||||
|
||||
```python
|
||||
# Basic usage
|
||||
answer = await pixelle_video.llm("What is machine learning?")
|
||||
|
||||
# With parameters
|
||||
answer = await pixelle_video.llm(
|
||||
prompt="Explain atomic habits",
|
||||
temperature=0.7, # 0.0-2.0 (lower = more deterministic)
|
||||
max_tokens=2000
|
||||
)
|
||||
```
|
||||
|
||||
### Environment Variables (Alternative)
|
||||
|
||||
Instead of `config.yaml`, you can use environment variables:
|
||||
|
||||
```bash
|
||||
# Qwen
|
||||
export DASHSCOPE_API_KEY="your-key"
|
||||
|
||||
# OpenAI
|
||||
export OPENAI_API_KEY="your-key"
|
||||
|
||||
# DeepSeek
|
||||
export DEEPSEEK_API_KEY="your-key"
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 2. TTS (Text-to-Speech)
|
||||
|
||||
### Configuration
|
||||
|
||||
Edit `config.yaml`:
|
||||
|
||||
```yaml
|
||||
tts:
|
||||
default: edge
|
||||
|
||||
edge:
|
||||
# No configuration needed - free to use!
|
||||
```
|
||||
|
||||
### Usage
|
||||
|
||||
```python
|
||||
# Basic usage (auto-generates temp path)
|
||||
audio_path = await pixelle_video.tts("Hello, world!")
|
||||
# Returns: "temp/abc123def456.mp3"
|
||||
|
||||
# With Chinese text
|
||||
audio_path = await pixelle_video.tts(
|
||||
text="你好，世界！",
|
||||
voice="zh-CN-YunjianNeural"
|
||||
)
|
||||
|
||||
# With custom parameters
|
||||
audio_path = await pixelle_video.tts(
|
||||
text="Welcome to Pixelle-Video",
|
||||
voice="en-US-JennyNeural",
|
||||
rate="+20%", # Speed: +50% = faster, -20% = slower
|
||||
volume="+0%",
|
||||
pitch="+0Hz"
|
||||
)
|
||||
|
||||
# Specify output path
|
||||
audio_path = await pixelle_video.tts(
|
||||
text="Hello",
|
||||
output_path="output/greeting.mp3"
|
||||
)
|
||||
```
|
||||
|
||||
### Popular Voices
|
||||
|
||||
**Chinese:**
|
||||
- `zh-CN-YunjianNeural` (male, default)
|
||||
- `zh-CN-XiaoxiaoNeural` (female)
|
||||
- `zh-CN-YunxiNeural` (male)
|
||||
- `zh-CN-XiaoyiNeural` (female)
|
||||
|
||||
**English:**
|
||||
- `en-US-JennyNeural` (female)
|
||||
- `en-US-GuyNeural` (male)
|
||||
- `en-GB-SoniaNeural` (female, British)
|
||||
|
||||
### List All Voices
|
||||
|
||||
```python
|
||||
# Get all available voices
|
||||
voices = await pixelle_video.tts.list_voices()
|
||||
|
||||
# Get Chinese voices only
|
||||
voices = await pixelle_video.tts.list_voices(locale="zh-CN")
|
||||
|
||||
# Get English voices only
|
||||
voices = await pixelle_video.tts.list_voices(locale="en-US")
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 3. Image Generation
|
||||
|
||||
### Configuration
|
||||
|
||||
Edit `config.yaml`:
|
||||
|
||||
```yaml
|
||||
image:
|
||||
default: comfykit
|
||||
|
||||
comfykit:
|
||||
# Local ComfyUI (optional, default: http://127.0.0.1:8188)
|
||||
comfyui_url: "http://127.0.0.1:8188"
|
||||
|
||||
# RunningHub cloud (optional)
|
||||
runninghub_api_key: "rh-key-xxx"
|
||||
```
|
||||
|
||||
### Usage
|
||||
|
||||
```python
|
||||
# Basic usage (local ComfyUI)
|
||||
image_url = await pixelle_video.image(
|
||||
workflow="workflows/book_cover_simple.json",
|
||||
prompt="minimalist book cover design, blue and white"
|
||||
)
|
||||
|
||||
# With full parameters
|
||||
image_url = await pixelle_video.image(
|
||||
workflow="workflows/book_cover_simple.json",
|
||||
prompt="book cover for 'Atomic Habits', professional, minimalist",
|
||||
negative_prompt="ugly, blurry, low quality",
|
||||
width=1024,
|
||||
height=1536,
|
||||
steps=20,
|
||||
seed=42
|
||||
)
|
||||
|
||||
# Using RunningHub cloud
|
||||
image_url = await pixelle_video.image(
|
||||
workflow="12345", # RunningHub workflow ID
|
||||
prompt="a beautiful landscape"
|
||||
)
|
||||
|
||||
# Check available workflows
|
||||
workflows = pixelle_video.image.list_workflows()
|
||||
print(f"Available workflows: {workflows}")
|
||||
```
|
||||
|
||||
### Environment Variables (Alternative)
|
||||
|
||||
```bash
|
||||
# Local ComfyUI
|
||||
export COMFYUI_BASE_URL="http://127.0.0.1:8188"
|
||||
|
||||
# RunningHub cloud
|
||||
export RUNNINGHUB_API_KEY="rh-key-xxx"
|
||||
```
|
||||
|
||||
### Workflow DSL
|
||||
|
||||
Pixelle-Video uses ComfyKit's DSL for workflow parameters:
|
||||
|
||||
```json
|
||||
{
|
||||
"6": {
|
||||
"class_type": "CLIPTextEncode",
|
||||
"_meta": {
|
||||
"title": "$prompt!"
|
||||
},
|
||||
"inputs": {
|
||||
"text": "default prompt",
|
||||
"clip": ["4", 1]
|
||||
}
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||
**DSL Markers:**
|
||||
- `$param!` - Required parameter
|
||||
- `$param` - Optional parameter
|
||||
- `$param~` - Upload parameter (for images/audio/video)
|
||||
- `$output.name` - Output variable
|
||||
|
||||
---
|
||||
|
||||
## Combined Workflow Example
|
||||
|
||||
Generate a complete book cover with narration:
|
||||
|
||||
```python
|
||||
import asyncio
|
||||
from pixelle_video.service import pixelle_video
|
||||
|
||||
async def create_book_content(book_title, author):
|
||||
"""Generate book summary, audio, and cover image"""
|
||||
|
||||
# 1. Generate book summary with LLM
|
||||
summary = await pixelle_video.llm(
|
||||
prompt=f"Write a compelling 2-sentence summary for a book titled '{book_title}' by {author}",
|
||||
max_tokens=100
|
||||
)
|
||||
print(f"Summary: {summary}")
|
||||
|
||||
# 2. Generate audio narration with TTS
|
||||
audio_path = await pixelle_video.tts(
|
||||
text=summary,
|
||||
voice="en-US-JennyNeural"
|
||||
)
|
||||
print(f"Audio: {audio_path}")
|
||||
|
||||
# 3. Generate book cover image
|
||||
image_url = await pixelle_video.image(
|
||||
workflow="workflows/book_cover_simple.json",
|
||||
prompt=f"book cover for '{book_title}' by {author}, professional, modern design",
|
||||
width=1024,
|
||||
height=1536
|
||||
)
|
||||
print(f"Cover: {image_url}")
|
||||
|
||||
return {
|
||||
"summary": summary,
|
||||
"audio": audio_path,
|
||||
"cover": image_url
|
||||
}
|
||||
|
||||
# Run
|
||||
result = asyncio.run(create_book_content("Atomic Habits", "James Clear"))
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Troubleshooting
|
||||
|
||||
### LLM Issues
|
||||
|
||||
**"API key not found"**
|
||||
- Make sure you've set the API key in `config.yaml` or environment variables
|
||||
- For Qwen: `DASHSCOPE_API_KEY`
|
||||
- For OpenAI: `OPENAI_API_KEY`
|
||||
- For DeepSeek: `DEEPSEEK_API_KEY`
|
||||
|
||||
**"Connection error"**
|
||||
- Check `base_url` in config
|
||||
- Verify API endpoint is accessible
|
||||
- For Ollama, make sure server is running (`ollama serve`)
|
||||
|
||||
### TTS Issues
|
||||
|
||||
**"SSL error"**
|
||||
- Edge TTS is free but requires internet connection
|
||||
- SSL verification is disabled by default for development
|
||||
|
||||
### Image Issues
|
||||
|
||||
**"ComfyUI connection refused"**
|
||||
- Make sure ComfyUI is running at http://127.0.0.1:8188
|
||||
- Or configure RunningHub API key for cloud execution
|
||||
|
||||
**"Workflow file not found"**
|
||||
- Check workflow path is correct
|
||||
- Use relative path from project root: `workflows/your_workflow.json`
|
||||
|
||||
**"No images generated"**
|
||||
- Check workflow has `SaveImage` node
|
||||
- Verify workflow parameters are correct
|
||||
- Check ComfyUI logs for errors
|
||||
|
||||
---
|
||||
|
||||
## Next Steps
|
||||
|
||||
- See `/examples/` directory for complete examples
|
||||
- Run `python test_integration.py` to test all capabilities
|
||||
- Create custom workflows in `/workflows/` directory
|
||||
- Check ComfyKit documentation: https://puke3615.github.io/ComfyKit
|
||||
|
||||
---
|
||||
|
||||
**Happy creating with Pixelle-Video!** 📚🎬
|
||||
|
||||
54
docs/en/development/architecture.md
Normal file
54
docs/en/development/architecture.md
Normal file
@@ -0,0 +1,54 @@
|
||||
# Architecture
|
||||
|
||||
Technical architecture overview of Pixelle-Video.
|
||||
|
||||
---
|
||||
|
||||
## Core Architecture
|
||||
|
||||
Pixelle-Video uses a layered architecture design:
|
||||
|
||||
- **Web Layer**: Streamlit Web interface
|
||||
- **Service Layer**: Core business logic
|
||||
- **ComfyUI Layer**: Image and TTS generation
|
||||
|
||||
---
|
||||
|
||||
## Main Components
|
||||
|
||||
### PixelleVideoCore
|
||||
|
||||
Core service class coordinating all sub-services.
|
||||
|
||||
### LLM Service
|
||||
|
||||
Responsible for calling large language models to generate scripts.
|
||||
|
||||
### Image Service
|
||||
|
||||
Responsible for calling ComfyUI to generate images.
|
||||
|
||||
### TTS Service
|
||||
|
||||
Responsible for calling ComfyUI to generate speech.
|
||||
|
||||
### Video Generator
|
||||
|
||||
Responsible for composing the final video.
|
||||
|
||||
---
|
||||
|
||||
## Tech Stack
|
||||
|
||||
- **Backend**: Python 3.10+, AsyncIO
|
||||
- **Web**: Streamlit
|
||||
- **AI**: OpenAI API, ComfyUI
|
||||
- **Configuration**: YAML
|
||||
- **Tools**: uv (package management)
|
||||
|
||||
---
|
||||
|
||||
## More Information
|
||||
|
||||
Detailed architecture documentation coming soon.
|
||||
|
||||
50
docs/en/development/contributing.md
Normal file
50
docs/en/development/contributing.md
Normal file
@@ -0,0 +1,50 @@
|
||||
# Contributing
|
||||
|
||||
Thank you for your interest in contributing to Pixelle-Video!
|
||||
|
||||
---
|
||||
|
||||
## How to Contribute
|
||||
|
||||
1. Fork the repository
|
||||
2. Create a feature branch (`git checkout -b feature/AmazingFeature`)
|
||||
3. Commit your changes (`git commit -m 'Add some AmazingFeature'`)
|
||||
4. Push to the branch (`git push origin feature/AmazingFeature`)
|
||||
5. Open a Pull Request
|
||||
|
||||
---
|
||||
|
||||
## Development Setup
|
||||
|
||||
```bash
|
||||
# Clone your fork
|
||||
git clone https://github.com/your-username/Pixelle-Video.git
|
||||
cd Pixelle-Video
|
||||
|
||||
# Install development dependencies
|
||||
uv sync
|
||||
|
||||
# Run tests
|
||||
pytest
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Code Standards
|
||||
|
||||
- All code and comments in English
|
||||
- Follow PEP 8 standards
|
||||
- Add appropriate tests
|
||||
|
||||
---
|
||||
|
||||
## Submit Issues
|
||||
|
||||
Having problems or feature suggestions? Please submit at [GitHub Issues](https://github.com/PixelleLab/Pixelle-Video/issues).
|
||||
|
||||
---
|
||||
|
||||
## Code of Conduct
|
||||
|
||||
Please be friendly and respectful. We are committed to fostering an inclusive community environment.
|
||||
|
||||
78
docs/en/faq.md
Normal file
78
docs/en/faq.md
Normal file
@@ -0,0 +1,78 @@
|
||||
# FAQ
|
||||
|
||||
Frequently Asked Questions.
|
||||
|
||||
---
|
||||
|
||||
## Installation
|
||||
|
||||
### Q: How to install uv?
|
||||
|
||||
```bash
|
||||
curl -LsSf https://astral.sh/uv/install.sh | sh
|
||||
```
|
||||
|
||||
### Q: Can I use something other than uv?
|
||||
|
||||
Yes, you can use traditional pip + venv approach.
|
||||
|
||||
---
|
||||
|
||||
## Configuration
|
||||
|
||||
### Q: Do I need to configure ComfyUI?
|
||||
|
||||
Not necessarily. You can use RunningHub cloud service without local deployment.
|
||||
|
||||
### Q: Which LLMs are supported?
|
||||
|
||||
All OpenAI-compatible LLMs, including:
|
||||
- Qianwen
|
||||
- GPT-4o
|
||||
- DeepSeek
|
||||
- Ollama (local)
|
||||
|
||||
---
|
||||
|
||||
## Usage
|
||||
|
||||
### Q: How long does first-time usage take?
|
||||
|
||||
Generating a 3-5 scene video takes approximately 2-5 minutes.
|
||||
|
||||
### Q: What if I'm not satisfied with the video?
|
||||
|
||||
Try:
|
||||
1. Change LLM model
|
||||
2. Adjust image dimensions and prompt prefix
|
||||
3. Change TTS workflow
|
||||
4. Try different video templates
|
||||
|
||||
### Q: What are the costs?
|
||||
|
||||
- **Completely Free**: Ollama + Local ComfyUI = $0
|
||||
- **Recommended**: Qianwen + Local ComfyUI ≈ $0.01-0.05/video
|
||||
- **Cloud Solution**: OpenAI + RunningHub (higher cost)
|
||||
|
||||
---
|
||||
|
||||
## Troubleshooting
|
||||
|
||||
### Q: ComfyUI connection failed
|
||||
|
||||
1. Confirm ComfyUI is running
|
||||
2. Check if URL is correct
|
||||
3. Click "Test Connection" in Web interface
|
||||
|
||||
### Q: LLM API call failed
|
||||
|
||||
1. Check if API Key is correct
|
||||
2. Check network connection
|
||||
3. Review error messages
|
||||
|
||||
---
|
||||
|
||||
## Other Questions
|
||||
|
||||
Have other questions? Check [Troubleshooting](troubleshooting.md) or submit an [Issue](https://github.com/PixelleLab/Pixelle-Video/issues).
|
||||
|
||||
45
docs/en/gallery/index.md
Normal file
45
docs/en/gallery/index.md
Normal file
@@ -0,0 +1,45 @@
|
||||
# 🎬 Video Gallery
|
||||
|
||||
Showcase of videos created with Pixelle-Video. Click on cards to view complete workflows and configuration files.
|
||||
|
||||
---
|
||||
|
||||
<div class="grid cards" markdown>
|
||||
|
||||
- **Reading Habit**
|
||||
|
||||
---
|
||||
|
||||
<video controls width="100%" style="border-radius: 8px;">
|
||||
<source src="https://your-oss-bucket.oss-cn-hangzhou.aliyuncs.com/pixelle-video/reading-habit/video.mp4" type="video/mp4">
|
||||
</video>
|
||||
|
||||
[:octicons-mark-github-16: View Workflows & Config](https://github.com/PixelleLab/Pixelle-Video/tree/main/docs/gallery/reading-habit)
|
||||
|
||||
- **Work Efficiency**
|
||||
|
||||
---
|
||||
|
||||
<video controls width="100%" style="border-radius: 8px;">
|
||||
<source src="https://your-oss-bucket.oss-cn-hangzhou.aliyuncs.com/pixelle-video/work-efficiency/video.mp4" type="video/mp4">
|
||||
</video>
|
||||
|
||||
[:octicons-mark-github-16: View Workflows & Config](https://github.com/PixelleLab/Pixelle-Video/tree/main/docs/gallery/work-efficiency)
|
||||
|
||||
- **Healthy Diet**
|
||||
|
||||
---
|
||||
|
||||
<video controls width="100%" style="border-radius: 8px;">
|
||||
<source src="https://your-oss-bucket.oss-cn-hangzhou.aliyuncs.com/pixelle-video/healthy-diet/video.mp4" type="video/mp4">
|
||||
</video>
|
||||
|
||||
[:octicons-mark-github-16: View Workflows & Config](https://github.com/PixelleLab/Pixelle-Video/tree/main/docs/gallery/healthy-diet)
|
||||
|
||||
</div>
|
||||
|
||||
---
|
||||
|
||||
!!! tip "How to Use"
|
||||
Click on a case card to jump to GitHub, download workflow files and configuration, and reproduce the video effect with one click.
|
||||
|
||||
60
docs/en/getting-started/configuration.md
Normal file
60
docs/en/getting-started/configuration.md
Normal file
@@ -0,0 +1,60 @@
|
||||
# Configuration
|
||||
|
||||
After installation, you need to configure services to use Pixelle-Video.
|
||||
|
||||
---
|
||||
|
||||
## LLM Configuration
|
||||
|
||||
LLM (Large Language Model) is used to generate video scripts.
|
||||
|
||||
### Quick Preset Selection
|
||||
|
||||
1. Select a preset model from the dropdown:
|
||||
- Qianwen (recommended, great value)
|
||||
- GPT-4o
|
||||
- DeepSeek
|
||||
- Ollama (local, completely free)
|
||||
|
||||
2. The system will auto-fill `base_url` and `model`
|
||||
|
||||
3. Click「🔑 Get API Key」to register and obtain credentials
|
||||
|
||||
4. Enter your API Key
|
||||
|
||||
---
|
||||
|
||||
## Image Configuration
|
||||
|
||||
Two options available:
|
||||
|
||||
### Local Deployment (Recommended)
|
||||
|
||||
Using local ComfyUI service:
|
||||
|
||||
1. Install and start ComfyUI
|
||||
2. Enter ComfyUI URL (default `http://127.0.0.1:8188`)
|
||||
3. Click "Test Connection" to verify
|
||||
|
||||
### Cloud Deployment
|
||||
|
||||
Using RunningHub cloud service:
|
||||
|
||||
1. Register for a RunningHub account
|
||||
2. Obtain API Key
|
||||
3. Enter API Key in configuration
|
||||
|
||||
---
|
||||
|
||||
## Save Configuration
|
||||
|
||||
After filling in all required configuration, click the "Save Configuration" button.
|
||||
|
||||
Configuration will be saved to `config.yaml` file.
|
||||
|
||||
---
|
||||
|
||||
## Next Steps
|
||||
|
||||
- [Quick Start](quick-start.md) - Create your first video
|
||||
|
||||
115
docs/en/getting-started/installation.md
Normal file
115
docs/en/getting-started/installation.md
Normal file
@@ -0,0 +1,115 @@
|
||||
# Installation
|
||||
|
||||
This page will guide you through installing Pixelle-Video.
|
||||
|
||||
---
|
||||
|
||||
## System Requirements
|
||||
|
||||
### Required
|
||||
|
||||
- **Python**: 3.10 or higher
|
||||
- **Operating System**: Windows, macOS, or Linux
|
||||
- **Package Manager**: uv (recommended) or pip
|
||||
|
||||
### Optional
|
||||
|
||||
- **GPU**: NVIDIA GPU with 6GB+ VRAM recommended for local ComfyUI
|
||||
- **Network**: Stable internet connection for LLM API and image generation services
|
||||
|
||||
---
|
||||
|
||||
## Installation Steps
|
||||
|
||||
### Step 1: Clone the Repository
|
||||
|
||||
```bash
|
||||
git clone https://github.com/PixelleLab/Pixelle-Video.git
|
||||
cd Pixelle-Video
|
||||
```
|
||||
|
||||
### Step 2: Install Dependencies
|
||||
|
||||
!!! tip "Recommended: Use uv"
|
||||
This project uses `uv` as the package manager, which is faster and more reliable than traditional pip.
|
||||
|
||||
#### Using uv (Recommended)
|
||||
|
||||
```bash
|
||||
# Install uv if you haven't already
|
||||
curl -LsSf https://astral.sh/uv/install.sh | sh
|
||||
|
||||
# Install project dependencies (uv will create a virtual environment automatically)
|
||||
uv sync
|
||||
```
|
||||
|
||||
#### Using pip
|
||||
|
||||
```bash
|
||||
# Create virtual environment
|
||||
python -m venv venv
|
||||
|
||||
# Activate virtual environment
|
||||
# Windows:
|
||||
venv\Scripts\activate
|
||||
# macOS/Linux:
|
||||
source venv/bin/activate
|
||||
|
||||
# Install dependencies
|
||||
pip install -e .
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Verify Installation
|
||||
|
||||
Run the following command to verify the installation:
|
||||
|
||||
```bash
|
||||
# Using uv
|
||||
uv run streamlit run web/app.py
|
||||
|
||||
# Or using pip (activate virtual environment first)
|
||||
streamlit run web/app.py
|
||||
```
|
||||
|
||||
Your browser should automatically open `http://localhost:8501` and display the Pixelle-Video web interface.
|
||||
|
||||
!!! success "Installation Successful!"
|
||||
If you can see the web interface, the installation was successful! Next, check out the [Configuration Guide](configuration.md) to set up your services.
|
||||
|
||||
---
|
||||
|
||||
## Optional: Install ComfyUI (Local Deployment)
|
||||
|
||||
If you want to run image generation locally, you'll need to install ComfyUI:
|
||||
|
||||
### Quick Install
|
||||
|
||||
```bash
|
||||
# Clone ComfyUI
|
||||
git clone https://github.com/comfyanonymous/ComfyUI.git
|
||||
cd ComfyUI
|
||||
|
||||
# Install dependencies
|
||||
pip install -r requirements.txt
|
||||
```
|
||||
|
||||
### Start ComfyUI
|
||||
|
||||
```bash
|
||||
python main.py
|
||||
```
|
||||
|
||||
ComfyUI runs on `http://127.0.0.1:8188` by default.
|
||||
|
||||
!!! info "ComfyUI Models"
|
||||
ComfyUI requires downloading model files to work. Please refer to the [ComfyUI documentation](https://github.com/comfyanonymous/ComfyUI) for information on downloading and configuring models.
|
||||
|
||||
---
|
||||
|
||||
## Next Steps
|
||||
|
||||
- [Configuration](configuration.md) - Configure LLM and image generation services
|
||||
- [Quick Start](quick-start.md) - Create your first video
|
||||
|
||||
107
docs/en/getting-started/quick-start.md
Normal file
107
docs/en/getting-started/quick-start.md
Normal file
@@ -0,0 +1,107 @@
|
||||
# Quick Start
|
||||
|
||||
Already installed and configured? Let's create your first video!
|
||||
|
||||
---
|
||||
|
||||
## Start the Web Interface
|
||||
|
||||
```bash
|
||||
# Using uv
|
||||
uv run streamlit run web/app.py
|
||||
```
|
||||
|
||||
Your browser will automatically open `http://localhost:8501`
|
||||
|
||||
---
|
||||
|
||||
## Create Your First Video
|
||||
|
||||
### Step 1: Check Configuration
|
||||
|
||||
On first use, expand the「⚙️ System Configuration」panel and confirm:
|
||||
|
||||
- **LLM Configuration**: Select an AI model (e.g., Qianwen, GPT) and enter API Key
|
||||
- **Image Configuration**: Configure ComfyUI address or RunningHub API Key
|
||||
|
||||
If not yet configured, see the [Configuration Guide](configuration.md).
|
||||
|
||||
Click "Save Configuration" when done.
|
||||
|
||||
---
|
||||
|
||||
### Step 2: Enter a Topic
|
||||
|
||||
In the left panel's「📝 Content Input」section:
|
||||
|
||||
1. Select「**AI Generate Content**」mode
|
||||
2. Enter a topic in the text box, for example:
|
||||
```
|
||||
Why develop a reading habit
|
||||
```
|
||||
3. (Optional) Set number of scenes, default is 5 frames
|
||||
|
||||
!!! tip "Topic Examples"
|
||||
- Why develop a reading habit
|
||||
- How to improve work efficiency
|
||||
- The importance of healthy eating
|
||||
- The meaning of travel
|
||||
|
||||
---
|
||||
|
||||
### Step 3: Configure Voice and Visuals
|
||||
|
||||
In the middle panel:
|
||||
|
||||
**Voice Settings**
|
||||
- Select TTS workflow (default Edge-TTS works well)
|
||||
- For voice cloning, upload a reference audio file
|
||||
|
||||
**Visual Settings**
|
||||
- Select image generation workflow (default works well)
|
||||
- Set image dimensions (default 1024x1024)
|
||||
- Choose video template (recommend portrait 1080x1920)
|
||||
|
||||
---
|
||||
|
||||
### Step 4: Generate Video
|
||||
|
||||
Click the「🎬 Generate Video」button in the right panel!
|
||||
|
||||
The system will show real-time progress:
|
||||
- Generate script
|
||||
- Generate images (for each scene)
|
||||
- Synthesize voice
|
||||
- Compose video
|
||||
|
||||
!!! info "Generation Time"
|
||||
Generating a 5-scene video takes about 2-5 minutes, depending on: LLM API response speed, image generation speed, TTS workflow type, and network conditions
|
||||
|
||||
---
|
||||
|
||||
### Step 5: Preview Video
|
||||
|
||||
Once complete, the video will automatically play in the right panel!
|
||||
|
||||
You'll see:
|
||||
- 📹 Video preview player
|
||||
- ⏱️ Video duration
|
||||
- 📦 File size
|
||||
- 🎬 Number of scenes
|
||||
- 📐 Video dimensions
|
||||
|
||||
The video file is saved in the `output/` folder.
|
||||
|
||||
---
|
||||
|
||||
## Next Steps
|
||||
|
||||
Congratulations! You've successfully created your first video 🎉
|
||||
|
||||
Next, you can:
|
||||
|
||||
- **Adjust Styles** - See the [Custom Visual Style](../tutorials/custom-style.md) tutorial
|
||||
- **Clone Voices** - See the [Voice Cloning with Reference Audio](../tutorials/voice-cloning.md) tutorial
|
||||
- **Use API** - See the [API Usage Guide](../user-guide/api.md)
|
||||
- **Develop Templates** - See the [Template Development Guide](../user-guide/templates.md)
|
||||
|
||||
97
docs/en/index.md
Normal file
97
docs/en/index.md
Normal file
@@ -0,0 +1,97 @@
|
||||
# Pixelle-Video 🎬
|
||||
|
||||
<div align="center" markdown="1">
|
||||
|
||||
**AI Video Creator - Generate a short video in 3 minutes**
|
||||
|
||||
[](https://github.com/PixelleLab/Pixelle-Video/stargazers)
|
||||
[](https://github.com/PixelleLab/Pixelle-Video/issues)
|
||||
[](https://github.com/PixelleLab/Pixelle-Video/blob/main/LICENSE)
|
||||
|
||||
</div>
|
||||
|
||||
---
|
||||
|
||||
## 🎯 Overview
|
||||
|
||||
Simply input a **topic**, and Pixelle-Video will automatically:
|
||||
|
||||
- ✍️ Write video scripts
|
||||
- 🎨 Generate AI images
|
||||
- 🗣️ Synthesize voice narration
|
||||
- 🎵 Add background music
|
||||
- 🎬 Create the final video
|
||||
|
||||
**No barriers, no video editing experience required** - turn video creation into a one-line task!
|
||||
|
||||
---
|
||||
|
||||
## ✨ Features
|
||||
|
||||
- ✅ **Fully Automated** - Input a topic, get a complete video in 3 minutes
|
||||
- ✅ **AI-Powered Scripts** - Intelligently create narration based on your topic
|
||||
- ✅ **AI-Generated Images** - Each sentence comes with beautiful AI illustrations
|
||||
- ✅ **AI Voice Synthesis** - Support for Edge-TTS, Index-TTS and more mainstream TTS solutions
|
||||
- ✅ **Background Music** - Add BGM for enhanced atmosphere
|
||||
- ✅ **Visual Styles** - Multiple templates to create unique video styles
|
||||
- ✅ **Flexible Dimensions** - Support for portrait, landscape and more video sizes
|
||||
- ✅ **Multiple AI Models** - Support for GPT, Qianwen, DeepSeek, Ollama, etc.
|
||||
- ✅ **Flexible Composition** - Based on ComfyUI architecture, use preset workflows or customize any capability
|
||||
|
||||
---
|
||||
|
||||
## 🎬 Video Examples
|
||||
|
||||
!!! info "Sample Videos"
|
||||
Coming soon: Video examples will be added here
|
||||
|
||||
---
|
||||
|
||||
## 🚀 Quick Start
|
||||
|
||||
Ready to get started? Just three steps:
|
||||
|
||||
1. **[Install Pixelle-Video](getting-started/installation.md)** - Download and install the project
|
||||
2. **[Configure Services](getting-started/configuration.md)** - Set up LLM and image generation services
|
||||
3. **[Create Your First Video](getting-started/quick-start.md)** - Start creating your first video
|
||||
|
||||
---
|
||||
|
||||
## 💰 Pricing
|
||||
|
||||
!!! success "Completely free to run!"
|
||||
|
||||
- **Completely Free**: Use Ollama (local) + Local ComfyUI = $0
|
||||
- **Recommended**: Use Qianwen LLM (≈$0.01-0.05 per 3-scene video) + Local ComfyUI
|
||||
- **Cloud Solution**: Use OpenAI + RunningHub (higher cost but no local setup required)
|
||||
|
||||
**Recommendation**: If you have a local GPU, go with the completely free solution. Otherwise, we recommend Qianwen for best value.
|
||||
|
||||
---
|
||||
|
||||
## 🤝 Acknowledgments
|
||||
|
||||
Pixelle-Video was inspired by the following excellent open source projects:
|
||||
|
||||
- [Pixelle-MCP](https://github.com/AIDC-AI/Pixelle-MCP) - ComfyUI MCP server
|
||||
- [MoneyPrinterTurbo](https://github.com/harry0703/MoneyPrinterTurbo) - Excellent video generation tool
|
||||
- [NarratoAI](https://github.com/linyqh/NarratoAI) - Video narration automation tool
|
||||
- [MoneyPrinterPlus](https://github.com/ddean2009/MoneyPrinterPlus) - Video creation platform
|
||||
- [ComfyKit](https://github.com/puke3615/ComfyKit) - ComfyUI workflow wrapper library
|
||||
|
||||
Thanks to these projects for their open source spirit! 🙏
|
||||
|
||||
---
|
||||
|
||||
## 📢 Feedback & Support
|
||||
|
||||
- 🐛 **Found a bug**: Submit an [Issue](https://github.com/PixelleLab/Pixelle-Video/issues)
|
||||
- 💡 **Feature request**: Submit a [Feature Request](https://github.com/PixelleLab/Pixelle-Video/issues)
|
||||
- ⭐ **Give us a Star**: If this project helps you, please give us a star!
|
||||
|
||||
---
|
||||
|
||||
## 📝 License
|
||||
|
||||
This project is licensed under the MIT License. See the [LICENSE](https://github.com/PixelleLab/Pixelle-Video/blob/main/LICENSE) file for details.
|
||||
|
||||
52
docs/en/reference/api-overview.md
Normal file
52
docs/en/reference/api-overview.md
Normal file
@@ -0,0 +1,52 @@
|
||||
# API Overview
|
||||
|
||||
Pixelle-Video Python API reference documentation.
|
||||
|
||||
---
|
||||
|
||||
## Core Classes
|
||||
|
||||
### PixelleVideoCore
|
||||
|
||||
Main service class providing video generation functionality.
|
||||
|
||||
```python
|
||||
from pixelle_video.service import PixelleVideoCore
|
||||
|
||||
pixelle = PixelleVideoCore()
|
||||
await pixelle.initialize()
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Main Methods
|
||||
|
||||
### generate_video()
|
||||
|
||||
Primary method for generating videos.
|
||||
|
||||
**Parameters**:
|
||||
|
||||
- `text` (str): Topic or complete script
|
||||
- `mode` (str): Generation mode ("generate" or "fixed")
|
||||
- `n_scenes` (int): Number of scenes
|
||||
- `title` (str, optional): Video title
|
||||
- `tts_workflow` (str): TTS workflow
|
||||
- `image_workflow` (str): Image generation workflow
|
||||
- `frame_template` (str): Video template
|
||||
- `bgm_path` (str, optional): BGM file path
|
||||
|
||||
**Returns**: `VideoResult` object
|
||||
|
||||
---
|
||||
|
||||
## Examples
|
||||
|
||||
Check the `examples/` directory for more examples.
|
||||
|
||||
---
|
||||
|
||||
## More Information
|
||||
|
||||
Detailed API documentation coming soon.
|
||||
|
||||
60
docs/en/reference/config-schema.md
Normal file
60
docs/en/reference/config-schema.md
Normal file
@@ -0,0 +1,60 @@
|
||||
# Config Schema
|
||||
|
||||
Detailed explanation of the `config.yaml` configuration file.
|
||||
|
||||
---
|
||||
|
||||
## Configuration Structure
|
||||
|
||||
```yaml
|
||||
llm:
|
||||
provider: openai
|
||||
api_key: "your-api-key"
|
||||
base_url: "https://dashscope.aliyuncs.com/compatible-mode/v1"
|
||||
model: "qwen-plus"
|
||||
|
||||
comfyui:
|
||||
comfyui_url: "http://127.0.0.1:8188"
|
||||
runninghub_api_key: ""
|
||||
|
||||
image:
|
||||
default_workflow: "runninghub/image_flux.json"
|
||||
prompt_prefix: "Minimalist illustration style"
|
||||
|
||||
tts:
|
||||
default_workflow: "selfhost/tts_edge.json"
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## LLM Configuration
|
||||
|
||||
- `provider`: Provider (currently only supports openai-compatible interfaces)
|
||||
- `api_key`: API key
|
||||
- `base_url`: API service address
|
||||
- `model`: Model name
|
||||
|
||||
---
|
||||
|
||||
## ComfyUI Configuration
|
||||
|
||||
### Basic Configuration
|
||||
|
||||
- `comfyui_url`: Local ComfyUI address
|
||||
- `runninghub_api_key`: RunningHub API key (optional)
|
||||
|
||||
### Image Configuration
|
||||
|
||||
- `default_workflow`: Default image generation workflow
|
||||
- `prompt_prefix`: Prompt prefix
|
||||
|
||||
### TTS Configuration
|
||||
|
||||
- `default_workflow`: Default TTS workflow
|
||||
|
||||
---
|
||||
|
||||
## More Information
|
||||
|
||||
The configuration file is automatically created on first run.
|
||||
|
||||
108
docs/en/troubleshooting.md
Normal file
108
docs/en/troubleshooting.md
Normal file
@@ -0,0 +1,108 @@
|
||||
# Troubleshooting
|
||||
|
||||
Having issues? Here are solutions to common problems.
|
||||
|
||||
---
|
||||
|
||||
## Installation Issues
|
||||
|
||||
### Dependency installation failed
|
||||
|
||||
```bash
|
||||
# Clean cache
|
||||
uv cache clean
|
||||
|
||||
# Reinstall
|
||||
uv sync
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Configuration Issues
|
||||
|
||||
### ComfyUI connection failed
|
||||
|
||||
**Possible Causes**:
|
||||
- ComfyUI not running
|
||||
- Incorrect URL configuration
|
||||
- Firewall blocking
|
||||
|
||||
**Solutions**:
|
||||
1. Confirm ComfyUI is running
|
||||
2. Check URL configuration (default `http://127.0.0.1:8188`)
|
||||
3. Test by accessing ComfyUI address in browser
|
||||
4. Check firewall settings
|
||||
|
||||
### LLM API call failed
|
||||
|
||||
**Possible Causes**:
|
||||
- Incorrect API Key
|
||||
- Network issues
|
||||
- Insufficient balance
|
||||
|
||||
**Solutions**:
|
||||
1. Verify API Key is correct
|
||||
2. Check network connection
|
||||
3. Review error message details
|
||||
4. Check account balance
|
||||
|
||||
---
|
||||
|
||||
## Generation Issues
|
||||
|
||||
### Video generation failed
|
||||
|
||||
**Possible Causes**:
|
||||
- Corrupted workflow file
|
||||
- Models not downloaded
|
||||
- Insufficient resources
|
||||
|
||||
**Solutions**:
|
||||
1. Check if workflow file exists
|
||||
2. Confirm ComfyUI has downloaded required models
|
||||
3. Check disk space and memory
|
||||
|
||||
### Image generation failed
|
||||
|
||||
**Solutions**:
|
||||
1. Check if ComfyUI is running properly
|
||||
2. Try manually testing workflow in ComfyUI
|
||||
3. Check workflow configuration
|
||||
|
||||
### TTS generation failed
|
||||
|
||||
**Solutions**:
|
||||
1. Check if TTS workflow is correct
|
||||
2. If using voice cloning, check reference audio format
|
||||
3. Review error logs
|
||||
|
||||
---
|
||||
|
||||
## Performance Issues
|
||||
|
||||
### Slow generation speed
|
||||
|
||||
**Optimization Tips**:
|
||||
1. Use local ComfyUI (faster than cloud)
|
||||
2. Reduce number of scenes
|
||||
3. Use faster LLM (e.g., Qianwen)
|
||||
4. Check network connection
|
||||
|
||||
---
|
||||
|
||||
## Other Issues
|
||||
|
||||
Still having problems?
|
||||
|
||||
1. Check project [GitHub Issues](https://github.com/PixelleLab/Pixelle-Video/issues)
|
||||
2. Submit a new Issue describing your problem
|
||||
3. Include error logs and configuration details for quick diagnosis
|
||||
|
||||
---
|
||||
|
||||
## View Logs
|
||||
|
||||
Log files are located in project root:
|
||||
- `api_server.log` - API service logs
|
||||
- `test_output.log` - Test logs
|
||||
|
||||
36
docs/en/tutorials/custom-style.md
Normal file
36
docs/en/tutorials/custom-style.md
Normal file
@@ -0,0 +1,36 @@
|
||||
# Custom Visual Style
|
||||
|
||||
Learn how to adjust image generation parameters to create unique visual styles.
|
||||
|
||||
---
|
||||
|
||||
## Adjust Prompt Prefix
|
||||
|
||||
The prompt prefix controls overall visual style:
|
||||
|
||||
```
|
||||
Minimalist black-and-white illustration, clean lines, simple style
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Adjust Image Dimensions
|
||||
|
||||
Different dimensions for different scenarios:
|
||||
|
||||
- **1024x1024**: Square, suitable for Xiaohongshu
|
||||
- **1080x1920**: Portrait, suitable for TikTok, Kuaishou
|
||||
- **1920x1080**: Landscape, suitable for Bilibili, YouTube
|
||||
|
||||
---
|
||||
|
||||
## Preview Effects
|
||||
|
||||
Use the "Preview Style" feature to test different configurations.
|
||||
|
||||
---
|
||||
|
||||
## More Information
|
||||
|
||||
More style customization tips coming soon.
|
||||
|
||||
35
docs/en/tutorials/voice-cloning.md
Normal file
35
docs/en/tutorials/voice-cloning.md
Normal file
@@ -0,0 +1,35 @@
|
||||
# Voice Cloning
|
||||
|
||||
Use reference audio to implement voice cloning functionality.
|
||||
|
||||
---
|
||||
|
||||
## Prepare Reference Audio
|
||||
|
||||
1. Prepare a clear audio file (MP3/WAV/FLAC)
|
||||
2. Recommended duration: 10-30 seconds
|
||||
3. Avoid background noise
|
||||
|
||||
---
|
||||
|
||||
## Usage Steps
|
||||
|
||||
1. Select a TTS workflow that supports voice cloning (e.g., Index-TTS) in voice settings
|
||||
2. Upload reference audio file
|
||||
3. Test effects with "Preview Voice"
|
||||
4. Generate video
|
||||
|
||||
---
|
||||
|
||||
## Notes
|
||||
|
||||
- Not all TTS workflows support voice cloning
|
||||
- Reference audio quality affects cloning results
|
||||
- Edge-TTS does not support voice cloning
|
||||
|
||||
---
|
||||
|
||||
## More Information
|
||||
|
||||
Detailed voice cloning tutorial coming soon.
|
||||
|
||||
33
docs/en/tutorials/your-first-video.md
Normal file
33
docs/en/tutorials/your-first-video.md
Normal file
@@ -0,0 +1,33 @@
|
||||
# Your First Video
|
||||
|
||||
Step-by-step guide to creating your first video with Pixelle-Video.
|
||||
|
||||
---
|
||||
|
||||
## Prerequisites
|
||||
|
||||
Make sure you've completed:
|
||||
|
||||
- ✅ [Installation](../getting-started/installation.md)
|
||||
- ✅ [Configuration](../getting-started/configuration.md)
|
||||
|
||||
---
|
||||
|
||||
## Tutorial Steps
|
||||
|
||||
For detailed steps, see [Quick Start](../getting-started/quick-start.md).
|
||||
|
||||
---
|
||||
|
||||
## Tips
|
||||
|
||||
- Choose an appropriate topic for better results
|
||||
- Start with 3-5 scenes for first generation
|
||||
- Preview voice and image effects before generating
|
||||
|
||||
---
|
||||
|
||||
## Troubleshooting
|
||||
|
||||
Having issues? Check out [FAQ](../faq.md) or [Troubleshooting](../troubleshooting.md).
|
||||
|
||||
42
docs/en/user-guide/api.md
Normal file
42
docs/en/user-guide/api.md
Normal file
@@ -0,0 +1,42 @@
|
||||
# API Usage
|
||||
|
||||
Pixelle-Video provides a complete Python API for easy integration into your projects.
|
||||
|
||||
---
|
||||
|
||||
## Quick Start
|
||||
|
||||
```python
|
||||
from pixelle_video.service import PixelleVideoCore
|
||||
import asyncio
|
||||
|
||||
async def main():
|
||||
# Initialize
|
||||
pixelle = PixelleVideoCore()
|
||||
await pixelle.initialize()
|
||||
|
||||
# Generate video
|
||||
result = await pixelle.generate_video(
|
||||
text="Why develop a reading habit",
|
||||
mode="generate",
|
||||
n_scenes=5
|
||||
)
|
||||
|
||||
print(f"Video generated: {result.video_path}")
|
||||
|
||||
# Run
|
||||
asyncio.run(main())
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## API Reference
|
||||
|
||||
For detailed API documentation, see [API Overview](../reference/api-overview.md).
|
||||
|
||||
---
|
||||
|
||||
## Examples
|
||||
|
||||
For more usage examples, check the `examples/` directory in the project.
|
||||
|
||||
48
docs/en/user-guide/templates.md
Normal file
48
docs/en/user-guide/templates.md
Normal file
@@ -0,0 +1,48 @@
|
||||
# Template Development
|
||||
|
||||
How to create custom video templates.
|
||||
|
||||
---
|
||||
|
||||
## Template Introduction
|
||||
|
||||
Video templates use HTML to define the layout and style of video frames.
|
||||
|
||||
---
|
||||
|
||||
## Template Structure
|
||||
|
||||
Templates are located in the `templates/` directory, grouped by size:
|
||||
|
||||
```
|
||||
templates/
|
||||
├── 1080x1920/ # Portrait
|
||||
├── 1920x1080/ # Landscape
|
||||
└── 1080x1080/ # Square
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## Creating Templates
|
||||
|
||||
1. Copy an existing template file
|
||||
2. Modify HTML and CSS
|
||||
3. Save to the corresponding size directory
|
||||
4. Select and use in Web interface
|
||||
|
||||
---
|
||||
|
||||
## Template Variables
|
||||
|
||||
Templates support the following variables:
|
||||
|
||||
- `{{ title }}` - Video title
|
||||
- `{{ text }}` - Scene text
|
||||
- `{{ image }}` - Scene image
|
||||
|
||||
---
|
||||
|
||||
## More Information
|
||||
|
||||
Detailed template development guide coming soon.
|
||||
|
||||
77
docs/en/user-guide/web-ui.md
Normal file
77
docs/en/user-guide/web-ui.md
Normal file
@@ -0,0 +1,77 @@
|
||||
# Web UI Guide
|
||||
|
||||
Detailed introduction to the Pixelle-Video Web interface features.
|
||||
|
||||
---
|
||||
|
||||
## Interface Layout
|
||||
|
||||
The Web interface uses a three-column layout:
|
||||
|
||||
- **Left Panel**: Content input and audio settings
|
||||
- **Middle Panel**: Voice and visual settings
|
||||
- **Right Panel**: Video generation and preview
|
||||
|
||||
---
|
||||
|
||||
## System Configuration
|
||||
|
||||
First-time use requires configuring LLM and image generation services. See [Configuration Guide](../getting-started/configuration.md).
|
||||
|
||||
---
|
||||
|
||||
## Content Input
|
||||
|
||||
### Generation Mode
|
||||
|
||||
- **AI Generate Content**: Enter a topic, AI creates script automatically
|
||||
- **Fixed Script Content**: Enter complete script directly
|
||||
|
||||
### Background Music
|
||||
|
||||
- Built-in music supported
|
||||
- Custom music files supported
|
||||
|
||||
---
|
||||
|
||||
## Voice Settings
|
||||
|
||||
### TTS Workflow
|
||||
|
||||
- Select TTS workflow
|
||||
- Supports Edge-TTS, Index-TTS, etc.
|
||||
|
||||
### Reference Audio
|
||||
|
||||
- Upload reference audio for voice cloning
|
||||
- Supports MP3/WAV/FLAC formats
|
||||
|
||||
---
|
||||
|
||||
## Visual Settings
|
||||
|
||||
### Image Generation
|
||||
|
||||
- Select image generation workflow
|
||||
- Set image dimensions
|
||||
- Adjust prompt prefix to control style
|
||||
|
||||
### Video Template
|
||||
|
||||
- Choose video template
|
||||
- Supports portrait/landscape/square
|
||||
- Preview template effects
|
||||
|
||||
---
|
||||
|
||||
## Generate Video
|
||||
|
||||
After clicking "Generate Video", the system will:
|
||||
|
||||
1. Generate video script
|
||||
2. Generate images for each scene
|
||||
3. Synthesize voice narration
|
||||
4. Compose final video
|
||||
|
||||
Automatically previews when complete.
|
||||
|
||||
37
docs/en/user-guide/workflows.md
Normal file
37
docs/en/user-guide/workflows.md
Normal file
@@ -0,0 +1,37 @@
|
||||
# Workflow Customization
|
||||
|
||||
How to customize ComfyUI workflows to achieve specific functionality.
|
||||
|
||||
---
|
||||
|
||||
## Workflow Introduction
|
||||
|
||||
Pixelle-Video is built on the ComfyUI architecture and supports custom workflows.
|
||||
|
||||
---
|
||||
|
||||
## Workflow Types
|
||||
|
||||
### TTS Workflows
|
||||
|
||||
Located in `workflows/selfhost/` or `workflows/runninghub/`
|
||||
|
||||
### Image Generation Workflows
|
||||
|
||||
Located in `workflows/selfhost/` or `workflows/runninghub/`
|
||||
|
||||
---
|
||||
|
||||
## Custom Workflows
|
||||
|
||||
1. Design your workflow in ComfyUI
|
||||
2. Export as JSON file
|
||||
3. Place in `workflows/` directory
|
||||
4. Select and use in Web interface
|
||||
|
||||
---
|
||||
|
||||
## More Information
|
||||
|
||||
Detailed workflow customization guide coming soon.
|
||||
|
||||
78
docs/gallery/index.md
Normal file
78
docs/gallery/index.md
Normal file
@@ -0,0 +1,78 @@
|
||||
# 🎬 视频示例库 / Video Gallery
|
||||
|
||||
展示使用 Pixelle-Video 制作的各类视频案例，包含完整的制作参数和资源下载。
|
||||
|
||||
Showcase of videos created with Pixelle-Video, including complete production parameters and downloadable resources.
|
||||
|
||||
---
|
||||
|
||||
## 📚 案例列表 / Cases
|
||||
|
||||
<div class="grid cards" markdown>
|
||||
|
||||
- :material-book-open-variant:{ .lg .middle } **阅读习惯养成**
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
**时长 Duration**: 45s | **分镜 Scenes**: 5 | **尺寸 Size**: 1080x1920
|
||||
|
||||
一个关于为什么要养成阅读习惯的教育科普视频。
|
||||
|
||||
An educational video about why we should develop reading habits.
|
||||
|
||||
[:octicons-arrow-right-24: 查看详情 View Details](reading-habit/)
|
||||
|
||||
- :material-chart-line:{ .lg .middle } **提高工作效率**
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
**时长 Duration**: 30s | **分镜 Scenes**: 3 | **尺寸 Size**: 1920x1080
|
||||
|
||||
关于如何提高工作效率的实用技巧分享。
|
||||
|
||||
Practical tips on improving work efficiency.
|
||||
|
||||
[:octicons-arrow-right-24: 查看详情 View Details](#) *(即将推出 Coming soon)*
|
||||
|
||||
- :material-food-apple:{ .lg .middle } **健康饮食**
|
||||
|
||||
---
|
||||
|
||||
|
||||
|
||||
**时长 Duration**: 60s | **分镜 Scenes**: 6 | **尺寸 Size**: 1080x1080
|
||||
|
||||
健康饮食的重要性和实用建议。
|
||||
|
||||
The importance of healthy eating and practical advice.
|
||||
|
||||
[:octicons-arrow-right-24: 查看详情 View Details](#) *(即将推出 Coming soon)*
|
||||
|
||||
</div>
|
||||
|
||||
---
|
||||
|
||||
## 🎯 如何使用这些案例 / How to Use
|
||||
|
||||
每个案例都包含：/ Each case includes:
|
||||
|
||||
- **📹 成品视频 Video**: OSS 托管的完整视频 / Complete video hosted on OSS
|
||||
- **⚙️ 工作流文件 Workflows**: ComfyUI 工作流 JSON / ComfyUI workflow JSON files
|
||||
- **📝 配置文件 Config**: 完整的生成配置 / Complete generation configuration
|
||||
- **🎨 提示词 Prompts**: 所有使用的提示词 / All prompts used
|
||||
- **📥 一键复现 Reproduce**: 可直接导入使用 / Can be imported directly
|
||||
|
||||
---
|
||||
|
||||
## 💡 贡献你的案例 / Contribute Your Case
|
||||
|
||||
制作了优秀的视频？欢迎分享！/ Created an awesome video? Share it with us!
|
||||
|
||||
查看 [贡献指南](../en/development/contributing.md) 了解如何提交你的案例。
|
||||
|
||||
See [Contributing Guide](../en/development/contributing.md) to learn how to submit your case.
|
||||
|
||||
1
docs/gallery/reading-habit/prompts.txt
Normal file
1
docs/gallery/reading-habit/prompts.txt
Normal file
@@ -0,0 +1 @@
|
||||
为什么要养成阅读习惯
|
||||
17
docs/stylesheets/extra.css
Normal file
17
docs/stylesheets/extra.css
Normal file
@@ -0,0 +1,17 @@
|
||||
/* Custom styles for Pixelle-Video documentation */
|
||||
|
||||
:root {
|
||||
--md-primary-fg-color: #5C6BC0;
|
||||
--md-accent-fg-color: #FF4081;
|
||||
}
|
||||
|
||||
/* Better code block styling */
|
||||
.highlight pre {
|
||||
border-radius: 4px;
|
||||
}
|
||||
|
||||
/* Admonition custom styling */
|
||||
.md-typeset .admonition {
|
||||
border-radius: 4px;
|
||||
}
|
||||
|
||||
54
docs/zh/development/architecture.md
Normal file
54
docs/zh/development/architecture.md
Normal file
@@ -0,0 +1,54 @@
|
||||
# 架构设计
|
||||
|
||||
Pixelle-Video 的技术架构概览。
|
||||
|
||||
---
|
||||
|
||||
## 核心架构
|
||||
|
||||
Pixelle-Video 采用分层架构设计：
|
||||
|
||||
- **Web 层**: Streamlit Web 界面
|
||||
- **服务层**: 核心业务逻辑
|
||||
- **ComfyUI 层**: 图像和TTS生成
|
||||
|
||||
---
|
||||
|
||||
## 主要组件
|
||||
|
||||
### PixelleVideoCore
|
||||
|
||||
核心服务类，协调各个子服务。
|
||||
|
||||
### LLM Service
|
||||
|
||||
负责调用大语言模型生成文案。
|
||||
|
||||
### Image Service
|
||||
|
||||
负责调用 ComfyUI 生成图像。
|
||||
|
||||
### TTS Service
|
||||
|
||||
负责调用 ComfyUI 生成语音。
|
||||
|
||||
### Video Generator
|
||||
|
||||
负责合成最终视频。
|
||||
|
||||
---
|
||||
|
||||
## 技术栈
|
||||
|
||||
- **后端**: Python 3.10+, AsyncIO
|
||||
- **Web**: Streamlit
|
||||
- **AI**: OpenAI API, ComfyUI
|
||||
- **配置**: YAML
|
||||
- **工具**: uv (包管理)
|
||||
|
||||
---
|
||||
|
||||
## 更多信息
|
||||
|
||||
详细的架构文档即将推出。
|
||||
|
||||
50
docs/zh/development/contributing.md
Normal file
50
docs/zh/development/contributing.md
Normal file
@@ -0,0 +1,50 @@
|
||||
# 贡献指南
|
||||
|
||||
感谢你对 Pixelle-Video 的贡献兴趣！
|
||||
|
||||
---
|
||||
|
||||
## 如何贡献
|
||||
|
||||
1. Fork 项目仓库
|
||||
2. 创建功能分支 (`git checkout -b feature/AmazingFeature`)
|
||||
3. 提交更改 (`git commit -m 'Add some AmazingFeature'`)
|
||||
4. 推送到分支 (`git push origin feature/AmazingFeature`)
|
||||
5. 开启 Pull Request
|
||||
|
||||
---
|
||||
|
||||
## 开发设置
|
||||
|
||||
```bash
|
||||
# 克隆你的 fork
|
||||
git clone https://github.com/your-username/Pixelle-Video.git
|
||||
cd Pixelle-Video
|
||||
|
||||
# 安装开发依赖
|
||||
uv sync
|
||||
|
||||
# 运行测试
|
||||
pytest
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 代码规范
|
||||
|
||||
- 所有代码和注释使用英文
|
||||
- 遵循 PEP 8 规范
|
||||
- 添加适当的测试
|
||||
|
||||
---
|
||||
|
||||
## 提交 Issue
|
||||
|
||||
遇到问题或有功能建议？请在 [GitHub Issues](https://github.com/PixelleLab/Pixelle-Video/issues) 提交。
|
||||
|
||||
---
|
||||
|
||||
## 行为准则
|
||||
|
||||
请保持友好和尊重，我们致力于营造包容的社区环境。
|
||||
|
||||
78
docs/zh/faq.md
Normal file
78
docs/zh/faq.md
Normal file
@@ -0,0 +1,78 @@
|
||||
# 常见问题
|
||||
|
||||
常见问题解答。
|
||||
|
||||
---
|
||||
|
||||
## 安装相关
|
||||
|
||||
### Q: 如何安装 uv？
|
||||
|
||||
```bash
|
||||
curl -LsSf https://astral.sh/uv/install.sh | sh
|
||||
```
|
||||
|
||||
### Q: 可以不用 uv 吗？
|
||||
|
||||
可以，你也可以使用传统的 pip + venv 方式。
|
||||
|
||||
---
|
||||
|
||||
## 配置相关
|
||||
|
||||
### Q: 必须要配置 ComfyUI 吗？
|
||||
|
||||
不一定。你可以使用 RunningHub 云端服务，无需本地部署。
|
||||
|
||||
### Q: 支持哪些 LLM？
|
||||
|
||||
支持所有 OpenAI 兼容接口的 LLM，包括：
|
||||
- 通义千问
|
||||
- GPT-4o
|
||||
- DeepSeek
|
||||
- Ollama（本地）
|
||||
|
||||
---
|
||||
|
||||
## 使用相关
|
||||
|
||||
### Q: 第一次使用需要多久？
|
||||
|
||||
生成一个 3-5 分镜的视频大约需要 2-5 分钟。
|
||||
|
||||
### Q: 视频效果不满意怎么办？
|
||||
|
||||
可以尝试：
|
||||
1. 更换 LLM 模型
|
||||
2. 调整图像尺寸和提示词前缀
|
||||
3. 更换 TTS 工作流
|
||||
4. 尝试不同的视频模板
|
||||
|
||||
### Q: 费用大概多少？
|
||||
|
||||
- **完全免费**: Ollama + 本地 ComfyUI = 0 元
|
||||
- **推荐方案**: 通义千问 + 本地 ComfyUI ≈ 0.01-0.05 元/视频
|
||||
- **云端方案**: OpenAI + RunningHub（费用较高）
|
||||
|
||||
---
|
||||
|
||||
## 故障排查
|
||||
|
||||
### Q: ComfyUI 连接失败
|
||||
|
||||
1. 确认 ComfyUI 正在运行
|
||||
2. 检查 URL 是否正确
|
||||
3. 在 Web 界面点击「测试连接」
|
||||
|
||||
### Q: LLM API 调用失败
|
||||
|
||||
1. 检查 API Key 是否正确
|
||||
2. 检查网络连接
|
||||
3. 查看错误提示
|
||||
|
||||
---
|
||||
|
||||
## 其他问题
|
||||
|
||||
有其他问题？请查看 [故障排查](troubleshooting.md) 或提交 [Issue](https://github.com/PixelleLab/Pixelle-Video/issues)。
|
||||
|
||||
45
docs/zh/gallery/index.md
Normal file
45
docs/zh/gallery/index.md
Normal file
@@ -0,0 +1,45 @@
|
||||
# 🎬 视频示例库
|
||||
|
||||
展示使用 Pixelle-Video 制作的视频案例。点击卡片查看完整的工作流和配置文件。
|
||||
|
||||
---
|
||||
|
||||
<div class="grid cards" markdown>
|
||||
|
||||
- **阅读习惯养成**
|
||||
|
||||
---
|
||||
|
||||
<video controls width="100%" style="border-radius: 8px;">
|
||||
<source src="https://your-oss-bucket.oss-cn-hangzhou.aliyuncs.com/pixelle-video/reading-habit/video.mp4" type="video/mp4">
|
||||
</video>
|
||||
|
||||
[:octicons-mark-github-16: 查看工作流和配置](https://github.com/PixelleLab/Pixelle-Video/tree/main/docs/gallery/reading-habit)
|
||||
|
||||
- **工作效率提升**
|
||||
|
||||
---
|
||||
|
||||
<video controls width="100%" style="border-radius: 8px;">
|
||||
<source src="https://your-oss-bucket.oss-cn-hangzhou.aliyuncs.com/pixelle-video/work-efficiency/video.mp4" type="video/mp4">
|
||||
</video>
|
||||
|
||||
[:octicons-mark-github-16: 查看工作流和配置](https://github.com/PixelleLab/Pixelle-Video/tree/main/docs/gallery/work-efficiency)
|
||||
|
||||
- **健康饮食**
|
||||
|
||||
---
|
||||
|
||||
<video controls width="100%" style="border-radius: 8px;">
|
||||
<source src="https://your-oss-bucket.oss-cn-hangzhou.aliyuncs.com/pixelle-video/healthy-diet/video.mp4" type="video/mp4">
|
||||
</video>
|
||||
|
||||
[:octicons-mark-github-16: 查看工作流和配置](https://github.com/PixelleLab/Pixelle-Video/tree/main/docs/gallery/healthy-diet)
|
||||
|
||||
</div>
|
||||
|
||||
---
|
||||
|
||||
!!! tip "如何使用"
|
||||
点击案例卡片跳转到 GitHub，下载工作流文件和配置，即可一键复现视频效果。
|
||||
|
||||
60
docs/zh/getting-started/configuration.md
Normal file
60
docs/zh/getting-started/configuration.md
Normal file
@@ -0,0 +1,60 @@
|
||||
# 配置说明
|
||||
|
||||
完成安装后，需要配置服务才能使用 Pixelle-Video。
|
||||
|
||||
---
|
||||
|
||||
## LLM 配置
|
||||
|
||||
LLM（大语言模型）用于生成视频文案。
|
||||
|
||||
### 快速选择预设
|
||||
|
||||
1. 从下拉菜单选择预设模型：
|
||||
- 通义千问（推荐，性价比高）
|
||||
- GPT-4o
|
||||
- DeepSeek
|
||||
- Ollama（本地运行，完全免费）
|
||||
|
||||
2. 系统会自动填充 `base_url` 和 `model`
|
||||
|
||||
3. 点击「🔑 获取 API Key」链接，注册并获取密钥
|
||||
|
||||
4. 填入 API Key
|
||||
|
||||
---
|
||||
|
||||
## 图像配置
|
||||
|
||||
支持两种方式：
|
||||
|
||||
### 本地部署（推荐）
|
||||
|
||||
使用本地 ComfyUI 服务：
|
||||
|
||||
1. 安装并启动 ComfyUI
|
||||
2. 填写 ComfyUI URL（默认 `http://127.0.0.1:8188`）
|
||||
3. 点击「测试连接」确认服务可用
|
||||
|
||||
### 云端部署
|
||||
|
||||
使用 RunningHub 云端服务：
|
||||
|
||||
1. 注册 RunningHub 账号
|
||||
2. 获取 API Key
|
||||
3. 在配置中填写 API Key
|
||||
|
||||
---
|
||||
|
||||
## 保存配置
|
||||
|
||||
填写完所有必需的配置后，点击「保存配置」按钮。
|
||||
|
||||
配置会保存到 `config.yaml` 文件中。
|
||||
|
||||
---
|
||||
|
||||
## 下一步
|
||||
|
||||
- [快速开始](quick-start.md) - 生成你的第一个视频
|
||||
|
||||
115
docs/zh/getting-started/installation.md
Normal file
115
docs/zh/getting-started/installation.md
Normal file
@@ -0,0 +1,115 @@
|
||||
# 安装
|
||||
|
||||
本页面将指导你完成 Pixelle-Video 的安装。
|
||||
|
||||
---
|
||||
|
||||
## 系统要求
|
||||
|
||||
### 必需条件
|
||||
|
||||
- **Python**: 3.10 或更高版本
|
||||
- **操作系统**: Windows、macOS 或 Linux
|
||||
- **包管理器**: uv（推荐）或 pip
|
||||
|
||||
### 可选条件
|
||||
|
||||
- **GPU**: 如需本地运行 ComfyUI，建议配备 NVIDIA 显卡（6GB+ 显存）
|
||||
- **网络**: 稳定的网络连接（用于调用 LLM API 和图像生成服务）
|
||||
|
||||
---
|
||||
|
||||
## 安装步骤
|
||||
|
||||
### 第一步：克隆项目
|
||||
|
||||
```bash
|
||||
git clone https://github.com/PixelleLab/Pixelle-Video.git
|
||||
cd Pixelle-Video
|
||||
```
|
||||
|
||||
### 第二步：安装依赖
|
||||
|
||||
!!! tip "推荐使用 uv"
|
||||
本项目使用 `uv` 作为包管理器，它比传统的 pip 更快、更可靠。
|
||||
|
||||
#### 使用 uv（推荐）
|
||||
|
||||
```bash
|
||||
# 如果还没有安装 uv，先安装它
|
||||
curl -LsSf https://astral.sh/uv/install.sh | sh
|
||||
|
||||
# 安装项目依赖（uv 会自动创建虚拟环境）
|
||||
uv sync
|
||||
```
|
||||
|
||||
#### 使用 pip
|
||||
|
||||
```bash
|
||||
# 创建虚拟环境
|
||||
python -m venv venv
|
||||
|
||||
# 激活虚拟环境
|
||||
# Windows:
|
||||
venv\Scripts\activate
|
||||
# macOS/Linux:
|
||||
source venv/bin/activate
|
||||
|
||||
# 安装依赖
|
||||
pip install -e .
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 验证安装
|
||||
|
||||
运行以下命令验证安装是否成功：
|
||||
|
||||
```bash
|
||||
# 使用 uv
|
||||
uv run streamlit run web/app.py
|
||||
|
||||
# 或使用 pip（需先激活虚拟环境）
|
||||
streamlit run web/app.py
|
||||
```
|
||||
|
||||
浏览器应该会自动打开 `http://localhost:8501`，显示 Pixelle-Video 的 Web 界面。
|
||||
|
||||
!!! success "安装成功！"
|
||||
如果能看到 Web 界面，说明安装成功了！接下来请查看 [配置说明](configuration.md) 来设置服务。
|
||||
|
||||
---
|
||||
|
||||
## 可选：安装 ComfyUI（本地部署）
|
||||
|
||||
如果希望本地运行图像生成服务，需要安装 ComfyUI：
|
||||
|
||||
### 快速安装
|
||||
|
||||
```bash
|
||||
# 克隆 ComfyUI
|
||||
git clone https://github.com/comfyanonymous/ComfyUI.git
|
||||
cd ComfyUI
|
||||
|
||||
# 安装依赖
|
||||
pip install -r requirements.txt
|
||||
```
|
||||
|
||||
### 启动 ComfyUI
|
||||
|
||||
```bash
|
||||
python main.py
|
||||
```
|
||||
|
||||
ComfyUI 默认运行在 `http://127.0.0.1:8188`
|
||||
|
||||
!!! info "ComfyUI 模型"
|
||||
ComfyUI 需要下载对应的模型文件才能工作。请参考 [ComfyUI 官方文档](https://github.com/comfyanonymous/ComfyUI) 了解如何下载和配置模型。
|
||||
|
||||
---
|
||||
|
||||
## 下一步
|
||||
|
||||
- [配置服务](configuration.md) - 配置 LLM 和图像生成服务
|
||||
- [快速开始](quick-start.md) - 生成第一个视频
|
||||
|
||||
107
docs/zh/getting-started/quick-start.md
Normal file
107
docs/zh/getting-started/quick-start.md
Normal file
@@ -0,0 +1,107 @@
|
||||
# 快速开始
|
||||
|
||||
已经完成安装和配置？让我们生成第一个视频吧！
|
||||
|
||||
---
|
||||
|
||||
## 启动 Web 界面
|
||||
|
||||
```bash
|
||||
# 使用 uv 运行
|
||||
uv run streamlit run web/app.py
|
||||
```
|
||||
|
||||
浏览器会自动打开 `http://localhost:8501`
|
||||
|
||||
---
|
||||
|
||||
## 生成你的第一个视频
|
||||
|
||||
### 步骤一：检查配置
|
||||
|
||||
首次使用时，展开「⚙️ 系统配置」面板，确认已配置：
|
||||
|
||||
- **LLM 配置**: 选择 AI 模型（如通义千问、GPT 等）并填入 API Key
|
||||
- **图像配置**: 配置 ComfyUI 地址或 RunningHub API Key
|
||||
|
||||
如果还没有配置，请查看 [配置说明](configuration.md)。
|
||||
|
||||
配置好后点击「保存配置」。
|
||||
|
||||
---
|
||||
|
||||
### 步骤二：输入主题
|
||||
|
||||
在左侧栏的「📝 内容输入」区域：
|
||||
|
||||
1. 选择「**AI 生成内容**」模式
|
||||
2. 在文本框中输入一个主题，例如：
|
||||
```
|
||||
为什么要养成阅读习惯
|
||||
```
|
||||
3. （可选）设置场景数量，默认 5 个分镜
|
||||
|
||||
!!! tip "主题示例"
|
||||
- 为什么要养成阅读习惯
|
||||
- 如何提高工作效率
|
||||
- 健康饮食的重要性
|
||||
- 旅行的意义
|
||||
|
||||
---
|
||||
|
||||
### 步骤三：配置语音和视觉
|
||||
|
||||
在中间栏：
|
||||
|
||||
**语音设置**
|
||||
- 选择 TTS 工作流（默认 Edge-TTS 即可）
|
||||
- 如需声音克隆，可上传参考音频
|
||||
|
||||
**视觉设置**
|
||||
- 选择图像生成工作流（默认即可）
|
||||
- 设置图像尺寸（默认 1024x1024）
|
||||
- 选择视频模板（推荐竖屏 1080x1920）
|
||||
|
||||
---
|
||||
|
||||
### 步骤四：生成视频
|
||||
|
||||
点击右侧栏的「🎬 生成视频」按钮！
|
||||
|
||||
系统会显示实时进度：
|
||||
- 生成文案
|
||||
- 生成配图（每个分镜）
|
||||
- 合成语音
|
||||
- 合成视频
|
||||
|
||||
!!! info "生成时间"
|
||||
生成一个 5 分镜的视频大约需要 2-5 分钟，具体时间取决于：LLM API 响应速度、图像生成速度、TTS 工作流类型、网络状况
|
||||
|
||||
---
|
||||
|
||||
### 步骤五：预览视频
|
||||
|
||||
生成完成后，视频会自动在右侧栏播放！
|
||||
|
||||
你可以看到：
|
||||
- 📹 视频预览播放器
|
||||
- ⏱️ 视频时长
|
||||
- 📦 文件大小
|
||||
- 🎬 分镜数量
|
||||
- 📐 视频尺寸
|
||||
|
||||
视频文件保存在 `output/` 文件夹中。
|
||||
|
||||
---
|
||||
|
||||
## 下一步探索
|
||||
|
||||
恭喜！你已经成功生成了第一个视频 🎉
|
||||
|
||||
接下来你可以：
|
||||
|
||||
- **调整风格** - 查看 [自定义视觉风格](../tutorials/custom-style.md) 教程
|
||||
- **克隆声音** - 查看 [使用参考音频克隆声音](../tutorials/voice-cloning.md) 教程
|
||||
- **使用 API** - 查看 [API 使用指南](../user-guide/api.md)
|
||||
- **开发模板** - 查看 [模板开发指南](../user-guide/templates.md)
|
||||
|
||||
97
docs/zh/index.md
Normal file
97
docs/zh/index.md
Normal file
@@ -0,0 +1,97 @@
|
||||
# Pixelle-Video 🎬
|
||||
|
||||
<div align="center" markdown="1">
|
||||
|
||||
**AI 视频创作工具 - 3 分钟生成一个短视频**
|
||||
|
||||
[](https://github.com/PixelleLab/Pixelle-Video/stargazers)
|
||||
[](https://github.com/PixelleLab/Pixelle-Video/issues)
|
||||
[](https://github.com/PixelleLab/Pixelle-Video/blob/main/LICENSE)
|
||||
|
||||
</div>
|
||||
|
||||
---
|
||||
|
||||
## 🎯 项目简介
|
||||
|
||||
只需输入一个 **主题**，Pixelle-Video 就能自动完成：
|
||||
|
||||
- ✍️ 撰写视频文案
|
||||
- 🎨 生成 AI 配图
|
||||
- 🗣️ 合成语音解说
|
||||
- 🎵 添加背景音乐
|
||||
- 🎬 一键合成视频
|
||||
|
||||
**零门槛，零剪辑经验**，让视频创作成为一句话的事！
|
||||
|
||||
---
|
||||
|
||||
## ✨ 功能亮点
|
||||
|
||||
- ✅ **全自动生成** - 输入主题，3 分钟自动生成完整视频
|
||||
- ✅ **AI 智能文案** - 根据主题智能创作解说词，无需自己写脚本
|
||||
- ✅ **AI 生成配图** - 每句话都配上精美的 AI 插图
|
||||
- ✅ **AI 生成语音** - 支持 Edge-TTS、Index-TTS 等众多主流 TTS 方案
|
||||
- ✅ **背景音乐** - 支持添加 BGM，让视频更有氛围
|
||||
- ✅ **视觉风格** - 多种模板可选，打造独特视频风格
|
||||
- ✅ **灵活尺寸** - 支持竖屏、横屏等多种视频尺寸
|
||||
- ✅ **多种 AI 模型** - 支持 GPT、通义千问、DeepSeek、Ollama 等
|
||||
- ✅ **原子能力灵活组合** - 基于 ComfyUI 架构，可使用预置工作流，也可自定义任意能力
|
||||
|
||||
---
|
||||
|
||||
## 🎬 视频示例
|
||||
|
||||
!!! info "示例视频"
|
||||
待补充：这里可以添加一些生成的视频示例
|
||||
|
||||
---
|
||||
|
||||
## 🚀 快速开始
|
||||
|
||||
想马上体验？只需三步：
|
||||
|
||||
1. **[安装 Pixelle-Video](getting-started/installation.md)** - 下载并安装项目
|
||||
2. **[配置服务](getting-started/configuration.md)** - 配置 LLM 和图像生成服务
|
||||
3. **[生成第一个视频](getting-started/quick-start.md)** - 开始创作你的第一个视频
|
||||
|
||||
---
|
||||
|
||||
## 💰 费用说明
|
||||
|
||||
!!! success "完全支持免费运行！"
|
||||
|
||||
- **完全免费方案**: LLM 使用 Ollama（本地运行）+ ComfyUI 本地部署 = 0 元
|
||||
- **推荐方案**: LLM 使用通义千问（生成一个 3 段视频约 0.01-0.05 元）+ ComfyUI 本地部署
|
||||
- **云端方案**: LLM 使用 OpenAI + 图像使用 RunningHub（费用较高但无需本地环境）
|
||||
|
||||
**选择建议**：本地有显卡建议完全免费方案，否则推荐使用通义千问（性价比高）
|
||||
|
||||
---
|
||||
|
||||
## 🤝 参考项目
|
||||
|
||||
Pixelle-Video 的设计受到以下优秀开源项目的启发：
|
||||
|
||||
- [Pixelle-MCP](https://github.com/AIDC-AI/Pixelle-MCP) - ComfyUI MCP 服务器，让 AI 助手直接调用 ComfyUI
|
||||
- [MoneyPrinterTurbo](https://github.com/harry0703/MoneyPrinterTurbo) - 优秀的视频生成工具
|
||||
- [NarratoAI](https://github.com/linyqh/NarratoAI) - 影视解说自动化工具
|
||||
- [MoneyPrinterPlus](https://github.com/ddean2009/MoneyPrinterPlus) - 视频创作平台
|
||||
- [ComfyKit](https://github.com/puke3615/ComfyKit) - ComfyUI 工作流封装库
|
||||
|
||||
感谢这些项目的开源精神！🙏
|
||||
|
||||
---
|
||||
|
||||
## 📢 反馈与支持
|
||||
|
||||
- 🐛 **遇到问题**: 提交 [Issue](https://github.com/PixelleLab/Pixelle-Video/issues)
|
||||
- 💡 **功能建议**: 提交 [Feature Request](https://github.com/PixelleLab/Pixelle-Video/issues)
|
||||
- ⭐ **给个 Star**: 如果这个项目对你有帮助，欢迎给个 Star 支持一下！
|
||||
|
||||
---
|
||||
|
||||
## 📝 许可证
|
||||
|
||||
本项目采用 MIT 许可证，详情请查看 [LICENSE](https://github.com/PixelleLab/Pixelle-Video/blob/main/LICENSE) 文件。
|
||||
|
||||
52
docs/zh/reference/api-overview.md
Normal file
52
docs/zh/reference/api-overview.md
Normal file
@@ -0,0 +1,52 @@
|
||||
# API 概览
|
||||
|
||||
Pixelle-Video Python API 参考文档。
|
||||
|
||||
---
|
||||
|
||||
## 核心类
|
||||
|
||||
### PixelleVideoCore
|
||||
|
||||
主要服务类，提供视频生成功能。
|
||||
|
||||
```python
|
||||
from pixelle_video.service import PixelleVideoCore
|
||||
|
||||
pixelle = PixelleVideoCore()
|
||||
await pixelle.initialize()
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 主要方法
|
||||
|
||||
### generate_video()
|
||||
|
||||
生成视频的主要方法。
|
||||
|
||||
**参数**:
|
||||
|
||||
- `text` (str): 主题或完整文案
|
||||
- `mode` (str): 生成模式 ("generate" 或 "fixed")
|
||||
- `n_scenes` (int): 分镜数量
|
||||
- `title` (str, optional): 视频标题
|
||||
- `tts_workflow` (str): TTS 工作流
|
||||
- `image_workflow` (str): 图像生成工作流
|
||||
- `frame_template` (str): 视频模板
|
||||
- `bgm_path` (str, optional): BGM 文件路径
|
||||
|
||||
**返回**: `VideoResult` 对象
|
||||
|
||||
---
|
||||
|
||||
## 示例
|
||||
|
||||
查看 `examples/` 目录获取更多示例。
|
||||
|
||||
---
|
||||
|
||||
## 更多信息
|
||||
|
||||
详细的 API 文档即将推出。
|
||||
|
||||
60
docs/zh/reference/config-schema.md
Normal file
60
docs/zh/reference/config-schema.md
Normal file
@@ -0,0 +1,60 @@
|
||||
# 配置文件详解
|
||||
|
||||
`config.yaml` 配置文件的详细说明。
|
||||
|
||||
---
|
||||
|
||||
## 配置结构
|
||||
|
||||
```yaml
|
||||
llm:
|
||||
provider: openai
|
||||
api_key: "your-api-key"
|
||||
base_url: "https://dashscope.aliyuncs.com/compatible-mode/v1"
|
||||
model: "qwen-plus"
|
||||
|
||||
comfyui:
|
||||
comfyui_url: "http://127.0.0.1:8188"
|
||||
runninghub_api_key: ""
|
||||
|
||||
image:
|
||||
default_workflow: "runninghub/image_flux.json"
|
||||
prompt_prefix: "Minimalist illustration style"
|
||||
|
||||
tts:
|
||||
default_workflow: "selfhost/tts_edge.json"
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## LLM 配置
|
||||
|
||||
- `provider`: 提供商（目前仅支持 openai 兼容接口）
|
||||
- `api_key`: API 密钥
|
||||
- `base_url`: API 服务地址
|
||||
- `model`: 模型名称
|
||||
|
||||
---
|
||||
|
||||
## ComfyUI 配置
|
||||
|
||||
### 基础配置
|
||||
|
||||
- `comfyui_url`: 本地 ComfyUI 地址
|
||||
- `runninghub_api_key`: RunningHub API 密钥（可选）
|
||||
|
||||
### 图像配置
|
||||
|
||||
- `default_workflow`: 默认图像生成工作流
|
||||
- `prompt_prefix`: 提示词前缀
|
||||
|
||||
### TTS 配置
|
||||
|
||||
- `default_workflow`: 默认 TTS 工作流
|
||||
|
||||
---
|
||||
|
||||
## 更多信息
|
||||
|
||||
配置文件会自动在首次运行时创建。
|
||||
|
||||
108
docs/zh/troubleshooting.md
Normal file
108
docs/zh/troubleshooting.md
Normal file
@@ -0,0 +1,108 @@
|
||||
# 故障排查
|
||||
|
||||
遇到问题？这里有一些常见问题的解决方案。
|
||||
|
||||
---
|
||||
|
||||
## 安装问题
|
||||
|
||||
### 依赖安装失败
|
||||
|
||||
```bash
|
||||
# 清理缓存
|
||||
uv cache clean
|
||||
|
||||
# 重新安装
|
||||
uv sync
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 配置问题
|
||||
|
||||
### ComfyUI 连接失败
|
||||
|
||||
**可能原因**:
|
||||
- ComfyUI 未运行
|
||||
- URL 配置错误
|
||||
- 防火墙阻止
|
||||
|
||||
**解决方案**:
|
||||
1. 确认 ComfyUI 正在运行
|
||||
2. 检查 URL 配置（默认 `http://127.0.0.1:8188`）
|
||||
3. 在浏览器中访问 ComfyUI 地址测试
|
||||
4. 检查防火墙设置
|
||||
|
||||
### LLM API 调用失败
|
||||
|
||||
**可能原因**:
|
||||
- API Key 错误
|
||||
- 网络问题
|
||||
- 余额不足
|
||||
|
||||
**解决方案**:
|
||||
1. 检查 API Key 是否正确
|
||||
2. 检查网络连接
|
||||
3. 查看错误提示中的具体原因
|
||||
4. 检查账户余额
|
||||
|
||||
---
|
||||
|
||||
## 生成问题
|
||||
|
||||
### 视频生成失败
|
||||
|
||||
**可能原因**:
|
||||
- 工作流文件损坏
|
||||
- 模型未下载
|
||||
- 资源不足
|
||||
|
||||
**解决方案**:
|
||||
1. 检查工作流文件是否存在
|
||||
2. 确认 ComfyUI 已下载所需模型
|
||||
3. 检查磁盘空间和内存
|
||||
|
||||
### 图像生成失败
|
||||
|
||||
**解决方案**:
|
||||
1. 检查 ComfyUI 是否正常运行
|
||||
2. 尝试在 ComfyUI 中手动测试工作流
|
||||
3. 检查工作流配置
|
||||
|
||||
### TTS 生成失败
|
||||
|
||||
**解决方案**:
|
||||
1. 检查 TTS 工作流是否正确
|
||||
2. 如使用声音克隆，检查参考音频格式
|
||||
3. 查看错误日志
|
||||
|
||||
---
|
||||
|
||||
## 性能问题
|
||||
|
||||
### 生成速度慢
|
||||
|
||||
**优化建议**:
|
||||
1. 使用本地 ComfyUI（比云端快）
|
||||
2. 减少分镜数量
|
||||
3. 使用更快的 LLM（如 Qianwen）
|
||||
4. 检查网络连接
|
||||
|
||||
---
|
||||
|
||||
## 其他问题
|
||||
|
||||
仍有问题？
|
||||
|
||||
1. 查看项目 [GitHub Issues](https://github.com/PixelleLab/Pixelle-Video/issues)
|
||||
2. 提交新的 Issue 描述你的问题
|
||||
3. 包含错误日志和配置信息以便快速定位
|
||||
|
||||
---
|
||||
|
||||
## 日志查看
|
||||
|
||||
日志文件位于项目根目录：
|
||||
- `api_server.log` - API 服务日志
|
||||
- `test_output.log` - 测试日志
|
||||
|
||||
36
docs/zh/tutorials/custom-style.md
Normal file
36
docs/zh/tutorials/custom-style.md
Normal file
@@ -0,0 +1,36 @@
|
||||
# 自定义视觉风格
|
||||
|
||||
学习如何调整图像生成参数以创建独特的视觉风格。
|
||||
|
||||
---
|
||||
|
||||
## 调整提示词前缀
|
||||
|
||||
提示词前缀控制整体视觉风格：
|
||||
|
||||
```
|
||||
Minimalist black-and-white illustration, clean lines, simple style
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 调整图像尺寸
|
||||
|
||||
不同尺寸适用于不同场景：
|
||||
|
||||
- **1024x1024**: 方形，适合小红书
|
||||
- **1080x1920**: 竖屏，适合抖音、快手
|
||||
- **1920x1080**: 横屏，适合B站、YouTube
|
||||
|
||||
---
|
||||
|
||||
## 预览效果
|
||||
|
||||
使用「预览风格」功能测试不同配置的效果。
|
||||
|
||||
---
|
||||
|
||||
## 更多信息
|
||||
|
||||
即将推出更多风格定制技巧。
|
||||
|
||||
35
docs/zh/tutorials/voice-cloning.md
Normal file
35
docs/zh/tutorials/voice-cloning.md
Normal file
@@ -0,0 +1,35 @@
|
||||
# 声音克隆
|
||||
|
||||
使用参考音频实现声音克隆功能。
|
||||
|
||||
---
|
||||
|
||||
## 准备参考音频
|
||||
|
||||
1. 准备一段清晰的音频文件（MP3/WAV/FLAC）
|
||||
2. 建议时长 10-30 秒
|
||||
3. 避免背景噪音
|
||||
|
||||
---
|
||||
|
||||
## 使用步骤
|
||||
|
||||
1. 在语音设置中选择支持声音克隆的 TTS 工作流（如 Index-TTS）
|
||||
2. 上传参考音频文件
|
||||
3. 使用「预览语音」测试效果
|
||||
4. 生成视频
|
||||
|
||||
---
|
||||
|
||||
## 注意事项
|
||||
|
||||
- 不是所有 TTS 工作流都支持声音克隆
|
||||
- 参考音频质量会影响克隆效果
|
||||
- Edge-TTS 不支持声音克隆
|
||||
|
||||
---
|
||||
|
||||
## 更多信息
|
||||
|
||||
即将推出更详细的声音克隆教程。
|
||||
|
||||
33
docs/zh/tutorials/your-first-video.md
Normal file
33
docs/zh/tutorials/your-first-video.md
Normal file
@@ -0,0 +1,33 @@
|
||||
# 生成你的第一个视频
|
||||
|
||||
手把手教你使用 Pixelle-Video 生成第一个视频。
|
||||
|
||||
---
|
||||
|
||||
## 前置准备
|
||||
|
||||
确保已完成：
|
||||
|
||||
- ✅ [安装](../getting-started/installation.md)
|
||||
- ✅ [配置](../getting-started/configuration.md)
|
||||
|
||||
---
|
||||
|
||||
## 教程步骤
|
||||
|
||||
详细步骤请查看 [快速开始](../getting-started/quick-start.md)。
|
||||
|
||||
---
|
||||
|
||||
## 小贴士
|
||||
|
||||
- 选择合适的主题可以获得更好的效果
|
||||
- 首次生成建议使用3-5个分镜
|
||||
- 可以先预览语音和图像效果
|
||||
|
||||
---
|
||||
|
||||
## 常见问题
|
||||
|
||||
遇到问题？查看 [FAQ](../faq.md) 或 [故障排查](../troubleshooting.md)。
|
||||
|
||||
42
docs/zh/user-guide/api.md
Normal file
42
docs/zh/user-guide/api.md
Normal file
@@ -0,0 +1,42 @@
|
||||
# API 使用
|
||||
|
||||
Pixelle-Video 提供完整的 Python API，方便集成到你的项目中。
|
||||
|
||||
---
|
||||
|
||||
## 快速开始
|
||||
|
||||
```python
|
||||
from pixelle_video.service import PixelleVideoCore
|
||||
import asyncio
|
||||
|
||||
async def main():
|
||||
# 初始化
|
||||
pixelle = PixelleVideoCore()
|
||||
await pixelle.initialize()
|
||||
|
||||
# 生成视频
|
||||
result = await pixelle.generate_video(
|
||||
text="为什么要养成阅读习惯",
|
||||
mode="generate",
|
||||
n_scenes=5
|
||||
)
|
||||
|
||||
print(f"视频已生成: {result.video_path}")
|
||||
|
||||
# 运行
|
||||
asyncio.run(main())
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## API 参考
|
||||
|
||||
详细 API 文档请查看 [API 概览](../reference/api-overview.md)。
|
||||
|
||||
---
|
||||
|
||||
## 示例
|
||||
|
||||
更多使用示例请参考项目的 `examples/` 目录。
|
||||
|
||||
48
docs/zh/user-guide/templates.md
Normal file
48
docs/zh/user-guide/templates.md
Normal file
@@ -0,0 +1,48 @@
|
||||
# 模板开发
|
||||
|
||||
如何创建自定义视频模板。
|
||||
|
||||
---
|
||||
|
||||
## 模板简介
|
||||
|
||||
视频模板使用 HTML 定义视频画面的布局和样式。
|
||||
|
||||
---
|
||||
|
||||
## 模板结构
|
||||
|
||||
模板位于 `templates/` 目录，按尺寸分组：
|
||||
|
||||
```
|
||||
templates/
|
||||
├── 1080x1920/ # 竖屏
|
||||
├── 1920x1080/ # 横屏
|
||||
└── 1080x1080/ # 方形
|
||||
```
|
||||
|
||||
---
|
||||
|
||||
## 创建模板
|
||||
|
||||
1. 复制现有模板文件
|
||||
2. 修改 HTML 和 CSS
|
||||
3. 保存到对应尺寸目录
|
||||
4. 在 Web 界面中选择使用
|
||||
|
||||
---
|
||||
|
||||
## 模板变量
|
||||
|
||||
模板支持以下变量：
|
||||
|
||||
- `{{ title }}` - 视频标题
|
||||
- `{{ text }}` - 分镜文本
|
||||
- `{{ image }}` - 分镜图片
|
||||
|
||||
---
|
||||
|
||||
## 更多信息
|
||||
|
||||
详细的模板开发指南即将推出。
|
||||
|
||||
77
docs/zh/user-guide/web-ui.md
Normal file
77
docs/zh/user-guide/web-ui.md
Normal file
@@ -0,0 +1,77 @@
|
||||
# Web 界面使用指南
|
||||
|
||||
详细介绍 Pixelle-Video Web 界面的各项功能。
|
||||
|
||||
---
|
||||
|
||||
## 界面布局
|
||||
|
||||
Web 界面采用三栏布局：
|
||||
|
||||
- **左侧栏**: 内容输入与音频设置
|
||||
- **中间栏**: 语音与视觉设置
|
||||
- **右侧栏**: 视频生成与预览
|
||||
|
||||
---
|
||||
|
||||
## 系统配置
|
||||
|
||||
首次使用需要配置 LLM 和图像生成服务。详见 [配置说明](../getting-started/configuration.md)。
|
||||
|
||||
---
|
||||
|
||||
## 内容输入
|
||||
|
||||
### 生成模式
|
||||
|
||||
- **AI 生成内容**: 输入主题，AI 自动创作文案
|
||||
- **固定文案内容**: 直接输入完整文案
|
||||
|
||||
### 背景音乐
|
||||
|
||||
- 支持内置音乐
|
||||
- 支持自定义音乐文件
|
||||
|
||||
---
|
||||
|
||||
## 语音设置
|
||||
|
||||
### TTS 工作流
|
||||
|
||||
- 选择 TTS 工作流
|
||||
- 支持 Edge-TTS、Index-TTS 等
|
||||
|
||||
### 参考音频
|
||||
|
||||
- 上传参考音频进行声音克隆
|
||||
- 支持 MP3/WAV/FLAC 等格式
|
||||
|
||||
---
|
||||
|
||||
## 视觉设置
|
||||
|
||||
### 图像生成
|
||||
|
||||
- 选择图像生成工作流
|
||||
- 设置图像尺寸
|
||||
- 调整提示词前缀控制风格
|
||||
|
||||
### 视频模板
|
||||
|
||||
- 选择视频模板
|
||||
- 支持竖屏/横屏/方形
|
||||
- 可预览模板效果
|
||||
|
||||
---
|
||||
|
||||
## 生成视频
|
||||
|
||||
点击「生成视频」按钮后，系统会：
|
||||
|
||||
1. 生成视频文案
|
||||
2. 为每个分镜生成配图
|
||||
3. 合成语音解说
|
||||
4. 合成最终视频
|
||||
|
||||
生成完成后自动预览。
|
||||
|
||||
37
docs/zh/user-guide/workflows.md
Normal file
37
docs/zh/user-guide/workflows.md
Normal file
@@ -0,0 +1,37 @@
|
||||
# 工作流定制
|
||||
|
||||
如何自定义 ComfyUI 工作流以实现特定功能。
|
||||
|
||||
---
|
||||
|
||||
## 工作流简介
|
||||
|
||||
Pixelle-Video 基于 ComfyUI 架构，支持自定义工作流。
|
||||
|
||||
---
|
||||
|
||||
## 工作流类型
|
||||
|
||||
### TTS 工作流
|
||||
|
||||
位于 `workflows/selfhost/` 或 `workflows/runninghub/`
|
||||
|
||||
### 图像生成工作流
|
||||
|
||||
位于 `workflows/selfhost/` 或 `workflows/runninghub/`
|
||||
|
||||
---
|
||||
|
||||
## 自定义工作流
|
||||
|
||||
1. 在 ComfyUI 中设计你的工作流
|
||||
2. 导出为 JSON 文件
|
||||
3. 放置到 `workflows/` 目录
|
||||
4. 在 Web 界面中选择使用
|
||||
|
||||
---
|
||||
|
||||
## 更多信息
|
||||
|
||||
即将推出更详细的工作流定制指南。
|
||||
|
||||
171
mkdocs.yml
Normal file
171
mkdocs.yml
Normal file
@@ -0,0 +1,171 @@
|
||||
site_name: Pixelle-Video
|
||||
site_description: AI Video Creator - Generate a short video in 3 minutes
|
||||
site_author: PixelleLab
|
||||
site_url: https://pixellelab.github.io/Pixelle-Video/
|
||||
|
||||
repo_name: PixelleLab/Pixelle-Video
|
||||
repo_url: https://github.com/PixelleLab/Pixelle-Video
|
||||
edit_uri: edit/main/docs/
|
||||
|
||||
copyright: Copyright © 2025 PixelleLab
|
||||
|
||||
theme:
|
||||
name: material
|
||||
language: en
|
||||
palette:
|
||||
# Light mode
|
||||
- media: "(prefers-color-scheme: light)"
|
||||
scheme: default
|
||||
primary: indigo
|
||||
accent: indigo
|
||||
toggle:
|
||||
icon: material/brightness-7
|
||||
name: Switch to dark mode
|
||||
# Dark mode
|
||||
- media: "(prefers-color-scheme: dark)"
|
||||
scheme: slate
|
||||
primary: indigo
|
||||
accent: indigo
|
||||
toggle:
|
||||
icon: material/brightness-4
|
||||
name: Switch to light mode
|
||||
|
||||
font:
|
||||
text: Roboto
|
||||
code: Roboto Mono
|
||||
|
||||
features:
|
||||
- navigation.instant # Instant loading
|
||||
- navigation.tracking # Anchor tracking
|
||||
- navigation.tabs # Top-level tabs
|
||||
- navigation.tabs.sticky # Sticky tabs
|
||||
- navigation.sections # Sidebar sections
|
||||
- navigation.expand # Expand sections
|
||||
- navigation.top # Back to top button
|
||||
- navigation.footer # Footer navigation
|
||||
- search.suggest # Search suggestions
|
||||
- search.highlight # Search highlighting
|
||||
- search.share # Share search results
|
||||
- content.code.copy # Copy button for code blocks
|
||||
- content.code.annotate # Code annotations
|
||||
- content.tabs.link # Link content tabs
|
||||
|
||||
icon:
|
||||
repo: fontawesome/brands/github
|
||||
|
||||
plugins:
|
||||
- search:
|
||||
lang:
|
||||
- en
|
||||
- zh
|
||||
- i18n:
|
||||
docs_structure: folder
|
||||
languages:
|
||||
- locale: en
|
||||
default: true
|
||||
name: English
|
||||
build: true
|
||||
- locale: zh
|
||||
name: 中文
|
||||
build: true
|
||||
nav_translations:
|
||||
Home: 首页
|
||||
Getting Started: 快速开始
|
||||
Installation: 安装
|
||||
Quick Start: 快速入门
|
||||
Configuration: 配置
|
||||
User Guide: 用户指南
|
||||
Web UI: Web 界面
|
||||
API Usage: API 使用
|
||||
Workflows: 工作流定制
|
||||
Templates: 模板开发
|
||||
Gallery: 示例库
|
||||
Tutorials: 教程
|
||||
Your First Video: 生成你的第一个视频
|
||||
Custom Style: 自定义视觉风格
|
||||
Voice Cloning: 声音克隆
|
||||
Reference: 参考
|
||||
API Overview: API 概览
|
||||
Config Schema: 配置文件详解
|
||||
Development: 开发指南
|
||||
Architecture: 架构设计
|
||||
Contributing: 贡献指南
|
||||
FAQ: 常见问题
|
||||
Troubleshooting: 故障排查
|
||||
- git-revision-date-localized:
|
||||
enable_creation_date: true
|
||||
type: datetime
|
||||
|
||||
markdown_extensions:
|
||||
# Python Markdown
|
||||
- abbr
|
||||
- admonition
|
||||
- attr_list
|
||||
- def_list
|
||||
- footnotes
|
||||
- md_in_html
|
||||
- toc:
|
||||
permalink: true
|
||||
|
||||
# Python Markdown Extensions
|
||||
- pymdownx.arithmatex:
|
||||
generic: true
|
||||
- pymdownx.betterem:
|
||||
smart_enable: all
|
||||
- pymdownx.caret
|
||||
- pymdownx.details
|
||||
- pymdownx.emoji:
|
||||
emoji_index: !!python/name:material.extensions.emoji.twemoji
|
||||
emoji_generator: !!python/name:material.extensions.emoji.to_svg
|
||||
- pymdownx.highlight:
|
||||
anchor_linenums: true
|
||||
line_spans: __span
|
||||
pygments_lang_class: true
|
||||
- pymdownx.inlinehilite
|
||||
- pymdownx.keys
|
||||
- pymdownx.mark
|
||||
- pymdownx.smartsymbols
|
||||
- pymdownx.superfences:
|
||||
custom_fences:
|
||||
- name: mermaid
|
||||
class: mermaid
|
||||
format: !!python/name:pymdownx.superfences.fence_code_format
|
||||
- pymdownx.tabbed:
|
||||
alternate_style: true
|
||||
- pymdownx.tasklist:
|
||||
custom_checkbox: true
|
||||
- pymdownx.tilde
|
||||
|
||||
nav:
|
||||
- Home: index.md
|
||||
- Getting Started:
|
||||
- Installation: getting-started/installation.md
|
||||
- Quick Start: getting-started/quick-start.md
|
||||
- Configuration: getting-started/configuration.md
|
||||
- User Guide:
|
||||
- Web UI: user-guide/web-ui.md
|
||||
- API Usage: user-guide/api.md
|
||||
- Workflows: user-guide/workflows.md
|
||||
- Templates: user-guide/templates.md
|
||||
- Gallery: gallery/index.md
|
||||
- Tutorials:
|
||||
- Your First Video: tutorials/your-first-video.md
|
||||
- Custom Style: tutorials/custom-style.md
|
||||
- Voice Cloning: tutorials/voice-cloning.md
|
||||
- Reference:
|
||||
- API Overview: reference/api-overview.md
|
||||
- Config Schema: reference/config-schema.md
|
||||
- Development:
|
||||
- Architecture: development/architecture.md
|
||||
- Contributing: development/contributing.md
|
||||
- FAQ: faq.md
|
||||
- Troubleshooting: troubleshooting.md
|
||||
|
||||
extra:
|
||||
social:
|
||||
- icon: fontawesome/brands/github
|
||||
link: https://github.com/PixelleLab/Pixelle-Video
|
||||
name: GitHub Repository
|
||||
|
||||
extra_css:
|
||||
- stylesheets/extra.css
|
||||
@@ -1,38 +0,0 @@
|
||||
"""
|
||||
Pixelle-Video CLI
|
||||
"""
|
||||
|
||||
import asyncio
|
||||
|
||||
from loguru import logger
|
||||
|
||||
from pixelle_video.service import pixelle_video
|
||||
|
||||
|
||||
async def test_llm():
|
||||
"""Test LLM capability"""
|
||||
# Initialize pixelle_video
|
||||
await pixelle_video.initialize()
|
||||
|
||||
# Test prompt
|
||||
prompt = "Explain the concept of atomic habits in 3 sentences."
|
||||
|
||||
logger.info(f"\n📝 Test Prompt: {prompt}\n")
|
||||
|
||||
# Call LLM
|
||||
result = await pixelle_video.llm(prompt)
|
||||
|
||||
logger.info(f"\n✨ Result:\n{result}\n")
|
||||
|
||||
|
||||
def main():
|
||||
"""Main CLI entry point"""
|
||||
logger.info("🚀 Pixelle-Video CLI\n")
|
||||
|
||||
# Run test
|
||||
asyncio.run(test_llm())
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
|
||||
Reference in New Issue
Block a user