README.md

metadata

title: Music Genre Classifier & Lyrics Generator
emoji: 🎵
colorFrom: indigo
colorTo: purple
sdk: gradio
sdk_version: 5.22.0
app_file: app.py
pinned: false
license: mit
short_description: AI music genre detection and lyrics generation

Music Analysis & Lyrics Generator

This Hugging Face Space application analyzes music files and generates lyrics that match the musical characteristics.

Features

• Music Analysis: Detects tempo, time signature, key, emotion, and theme
• Genre Classification: Identifies the music genre using a pre-trained classifier
• Lyrics Generation: Creates lyrics that match the style, emotion, and length of your music using Qwen3-32B

How to Use

1. Upload a music file or record audio directly in the app
2. Click "Analyze and Generate Lyrics"
3. View the analysis results showing tempo, key, emotion, theme, and genre
4. Check the generated lyrics tailored to match your music

Supported Genres

Lyrics generation is currently limited to the following genres:

• Pop
• Rock
• Country
• Disco
• Metal

These genres have consistent syllable-to-beat relationships that work well with our beat-matching algorithm. For other genres, the application will still provide music analysis, but lyrics generation will be disabled.

Technical Details

This application uses:

• MusicAnalyzer: Custom analysis tool for detecting musical features
• Hugging Face Transformers: Pre-trained models for genre classification and lyrics generation
• Gradio: For the user interface
• Librosa: For audio processing

Requirements

See requirements.txt for detailed dependencies.

Limitations

• Large audio files may take longer to process
• The quality of lyrics generation depends on the clarity of the audio and the detected musical features
• Lyrics generation is restricted to specific genres (see Supported Genres section)

Credits

• Genre classification model: dima806/music_genres_classification
• LLM for lyrics generation: Qwen/Qwen3-32B