Describe the music you want to create

Duration (seconds)

5 30

Seed (leave empty for random)

Try These Examples

Describe the music you want to create	Duration (seconds)	Seed (leave empty for random)	Temperature (higher = more creative, lower = more focused)	Top-k sampling	Top-p sampling (0 to disable)	Guidance Scale (how closely to follow the prompt)	Genre	Mood	Featured Instruments	Tempo

Generated Music

Download Generated Music

Audio Analysis

Tempo: -

Brightness: -

Duration: -

Recent Generations

Generated	Prompt

Replay Generation

Selected Generation

Download Selected Generation

Parameters Used

About MusicGen Text-to-Music Generator

This app uses Meta's MusicGen model to generate music from text descriptions. The model is based on a 1.5 billion parameter neural network trained on music and text pairs.

How It Works

Enter a text prompt describing the music you want to create
Adjust parameters like duration, temperature, and other settings
Click Generate and watch the progress bar as your music is created
Listen and download your generated music

Tips for Better Results

Be specific in your descriptions
Include details about genre, instruments, mood, and tempo
Use adjectives to describe the sound you want
Experiment with different parameters:
- Higher temperature = more creative but potentially chaotic
- Lower temperature = more predictable but may be repetitive
- Guidance scale controls how closely the model follows your prompt

Model Information

MusicGen is a text-to-music model capable of generating high-quality music samples conditioned on text descriptions. It was developed by Meta AI and published in the paper "Simple and Controllable Music Generation."

Model size: 1.5 billion parameters (medium version)
Sampling rate: 32kHz
License: CC-BY-NC 4.0

Learn more about MusicGen

Created By

This Hugging Face Space was created by Lalo - Software Developer & Data Engineer.

Check out more projects on my YouTube channel!

🎵 MusicGen: Text-to-Music Generator