Skip to content

Dataset · vikhyatk

7,000+ Hours of Generated Lofi Music

vikhyatk/lofi

An audio corpus of 7,000+ hours of lofi music with text prompts, generated by MusicGen Large.

What's inside

The dataset contains over 7,000 hours of lofi music generated by the MusicGen Large model using diverse prompts. Prompts were generated by Llama 3.1 8B Base starting from a set of 1,960 handwritten prompts; 16 random ones were used in few-shot mode. Modalities are audio and text; CC-BY-NC license.

What you can do with it

  • Training a text-to-music model in the lofi style
  • Fine-tuning MusicGen or LoRA for musical prompts
  • Evaluating generative audio systems on long tracks
Size
100K–1M rows
Downloads this month
3.1K
Likes
88

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("vikhyatk/lofi")