# Can MusicGen Create Training Data for MIR Tasks?

arXiv: https://arxiv.org/abs/2311.09094 · HTML: https://mlsanity.com/paper/2311.09094/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 9 references that are in the map.
- First posted on arXiv 15 Nov 2023.

## Builds on (most cited first, 6 of 6)

- [Simple and Controllable Music Generation](https://mlsanity.com/paper/2306.05284/) (2023)
- [MusicLM: Generating Music From Text](https://mlsanity.com/paper/2301.11325/) (2023)
- [A Brief Review of Domain Adaptation](https://mlsanity.com/paper/2010.03978/) (2020)
- [Efficient Training of Audio Transformers with Patchout](https://mlsanity.com/paper/2110.05069/) (2021)
- [JEN-1: Text-Guided Universal Music Generation with Omnidirectional Diffusion Models](https://mlsanity.com/paper/2308.04729/) (2023)
- [Audio Embeddings as Teachers for Music Classification](https://mlsanity.com/paper/2306.17424/) (2023)

## Built on by (most cited first, 1 of 1)

- [Towards Training Music Taggers on Synthetic Data](https://mlsanity.com/paper/2407.02156/) (2024)

## Abstract (arXiv)

We are investigating the broader concept of using AI-based generative music systems to generate training data for Music Information Retrieval (MIR) tasks. To kick off this line of work, we ran an initial experiment in which we trained a genre classifier on a fully artificial music dataset created with MusicGen. We constructed over 50 000 genre- conditioned textual descriptions and generated a collection of music excerpts that covers five musical genres. Our preliminary results show that the proposed model can learn genre-specific characteristics from artificial music tracks that generalise well to real-world music recordings.