A ComfyUI custom node for generating dynamic scrolling lyrics videos with smooth animations. Supports audio-to-text transcription using OpenAI Whisper or importing existing SRT subtitles.

一个用于生成动态滚动歌词视频的 ComfyUI 自定义节点。支持使用 OpenAI Whisper 进行语音转文字，或导入现有的 SRT 字幕。

工作流1

工作流2

效果演示

工作流下载

🔗 GitHub Repository: https://github.com/ahkimkoo/Comfyui-Lyrics

✨ Features (功能特性)

Dynamic Scrolling (动态滚动):
- Smooth slot-based scrolling animation with Ease-Out cubic curves.
- Active line magnification and highlighting.
- Automatic "Upcoming" preview logic.
- 基于槽位的平滑滚动动画，支持焦点行放大高亮，上下行半透明预览。
Audio Transcription (音频识别):
- Integrated OpenAI Whisper for automatic speech-to-text.
- Support for initial_prompt to guide language style (e.g., Simplified/Traditional Chinese).
- 集成 OpenAI Whisper，支持自动语音识别生成字幕。支持设置 Prompt 以引导简繁体输出。
SRT Support (字幕支持):
- Import existing SRT text directly.
- Auto-fallback logic: Use text if provided, otherwise transcribe audio.
- 支持直接输入 SRT 格式文本。智能判断：优先使用文本，若为空则自动识别音频。
Rich Styling (丰富样式):
- Custom Fonts (support for models/fonts).
- Text Shadow & Stroke (阴影与描边).
- Adjustable Margins & Spacing (边距与行间距).
- Dynamic Triangle Indicator (动态三角指示标).
Smart Layout (智能排版):
- Automatic text wrapping (自动折行).
- Dynamic height calculation based on content.
- Aligned indentation for list-style look.

📦 Installation (安装)

Method 1: ComfyUI Manager (Recommended)

You can install this node directly via ComfyUI Manager:

Open ComfyUI Manager.
Click "Install Custom Nodes".
Search for Comfyui-Lyrics.
Install.

在 ComfyUI Manager 中搜索 Comfyui-Lyrics 即可一键安装。

Method 2: Manual Installation

Navigate to your ComfyUI custom nodes directory:
```
cd ComfyUI/custom_nodes/
```

Clone this repository:

git clone https://github.com/ahkimkoo/Comfyui-Lyrics.git

Install dependencies:
```
cd Comfyui-Lyrics
pip install -r requirements.txt
```
Dependencies include: openai-whisper, Pillow, torchaudio, numpy, torch.
Optional: Place your font files (.ttf, .otf) in ComfyUI/models/fonts/. If the directory doesn't exist, create it.
- 提示: 请在 ComfyUI/models/fonts/ 下放入中文字体文件（如 msyh.ttf），否则中文可能无法正确显示。

🚀 Usage (使用方法)

Add the node "Lyrics Scroll Effect" (Category: Lyrics) to your workflow.

Inputs (输入参数)

width / height: Video resolution (视频分辨率).
margin_left / margin_right: Side margins (左右边距). Text width is calculated automatically.
y_pos: Y-position of the Active Line (Center). (焦点行垂直位置).
font_size: Font size for inactive lines (非焦点行字号). Default: 30.
active_font_size: Font size for the active line (焦点行字号). Default: 40.
letter_spacing: Spacing between characters (字间距).
line_gap: Spacing between lines/items (行间距/项间距). Also applies to wrapped lines.
text_color: Hex color for text (文字颜色).
stroke_width / stroke_color: Text outline settings (描边设置).
shadow_color / shadow_offset / shadow_alpha: Drop shadow settings (阴影设置).
font_filename: Select font from models/fonts.
frame_rate: Output video frame rate (e.g., 25.0).
whisper_prompt: Hint for Whisper (e.g., "简体中文", "English"). Controls language style.
audio (Optional): Audio input for timing/transcription.
text (Optional): SRT format text.

Logic Priority (逻辑优先级)

Text Input: If text is connected and not empty, it uses the provided SRT text. Audio is used for duration calculation only.
Audio Input: If text is empty, it uses audio to run Whisper transcription.
Duration: Calculated from Audio length (if present) or Subtitle length (if Text only).

📄 License

This project is open source.

Created by ahkimkoo