ComfyUI-Mzikart-Singer
About this model
Demo songs produced using this workflow
# ComfyUI-Mzikart-Singer
A comprehensive ComfyUI node pack for AI music generation with advanced lyrics integration and genre-specific optimization.
## ๐ต Features
- **Genre-Specific Prompts**: 25+ music genres with tailored positive prompts
- **Advanced Negative Filtering**: Genre-specific negative tags to avoid unwanted elements
- **Lyrics Integration**: Load, clean, and integrate lyrics with structure preservation
- **Vocal Customization**: 20+ vocal styles and gender options
- **File Organization**: Recursive lyrics loading from organized folder structures๐ฆ Installation
Clone this repository into your ComfyUI custom nodes directory:
cd ComfyUI/custom_nodes/
git clone ComfyUI๐ค Nodes Overview
๐ต Mzikart Negative Cleaner
Generates optimized negative prompts to filter out unwanted audio characteristics.
Inputs:
genre: Select from Universal, Trap, Gospel, R&B, Hip Hop
additional_tags: Custom negative tags (optional)
Outputs:
negative_tags: Combined negative prompt string
๐ค Mzikart Lyrics Generator
Creates comprehensive positive prompts with lyrics integration and vocal styling.
Inputs:
genre: 25+ genres including Trap, Gospel, R&B, Jazz, Rock, etc.
singer_gender: male, female, mixed, choir, none
vocal_style: 20+ styles (raspy, breathy, powerful, etc.)
bpm: Tempo (60-200 BPM)
mood: neutral, happy, sad, aggressive, romantic, mysterious
lyrics: Direct lyrics input (multiline)
additional_tags: Custom positive tags (optional)
Outputs:
positive_tags: Complete positive prompt
full_lyrics: Cleaned and formatted lyrics
๐ Mzikart Lyrics Loader
Loads lyrics from .txt files with support for organized folder structures.
Inputs:
lyrics_file: Select from available .txt files
Outputs:
lyrics: Loaded lyrics content
๐ File Organization
The lyrics loader automatically creates a lyrics folder in your ComfyUI models directory and supports nested organization:
ComfyUI/models/checkpoints/lyrics/
โโโ song1.txt
โโโ artist1/
โ โโโ album1/
โ โ โโโ track1.txt
โ โโโ singles/
โ โโโ single1.txt
โโโ genres/
โโโ trap/
โ โโโ trap_song.txt
โโโ gospel/
โโโ gospel_song.txt
๐ผ Supported Genres
Main Categories:
Trap, Trap Melodic, Trap Gospel
Gospel, R&B, Hip Hop
Jazz, Soul, Blues
Rock, Metal, Pop
Electronic, EDM, Dubstep
Country, Folk, Classical
And 15+ more specialized genres
๐๏ธ Vocal Styles
Basic: mainstream, soft, powerful
Textural: raspy, breathy, gravelly
Technical: melismatic, vibrato, falsetto
Stylistic: opera, crooning, scat, spoken-rap
Effects: robotic, whispered, shouting
๐ง Usage Examples
Basic Workflow
Use Mzikart Lyrics Loader to load lyrics from file
Connect to Mzikart Lyrics Generator for positive prompts
Use Mzikart Negative Cleaner for negative prompts
Feed both to your AI music generation model
Advanced Configuration
Genre: Trap Melodic
Singer: female
Vocal Style: powerful
BPM: 140
Mood: aggressive
๐ ๏ธ Technical Features
Lyrics Cleaning: Preserves apostrophes, punctuation, and structure
Cross-Platform: Works on Windows, macOS, and Linux
Unicode Support: Handles international characters
Error Handling: Graceful fallbacks for missing files
Tags
Related Models
Similar AI models you may like
ON-THE-FLY ๅฎๆถ็ๆ๏ผWan-AI ไธ็ธ/ Wan2.1 Video Model (multi-specs) - CausVid&Comfy&Kijai - workflow included
ใWAN2.1ใIMG to VIDEO
ComfyUI Image Workflows
WAN 2.2 Workflow T2V-I2V-T2I (Kijai Wrapper)
Hand Detailer/Segmentation - ADetailer
Hunyuan ๐ป AllInOne
Moody Simple Zimage Turbo/Distilled Workflow