# Enriching Music Descriptions with a Finetuned-LLM and Metadata for Text-to-Music Retrieval

arXiv: https://arxiv.org/abs/2410.03264 · HTML: https://mlsanity.com/paper/2410.03264/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Oct 2024.

## Built on by (most cited first, 7 of 7)

- [CLaMP 2: Multimodal Music Information Retrieval Across 101 Languages Using Large Language Models](https://mlsanity.com/paper/2410.13267/) (2024)
- [DanceChat: Large Language Model-Guided Music-to-Dance Generation](https://mlsanity.com/paper/2506.10574/) (2025)
- [PIAST: A Multimodal Piano Dataset with Audio, Symbolic and Text](https://mlsanity.com/paper/2411.02551/) (2024)
- [Rethinking Music Captioning with Music Metadata LLMs](https://mlsanity.com/paper/2602.03023/) (2026)
- [Music Recommendation with Large Language Models: Challenges, Opportunities, and Evaluation](https://mlsanity.com/paper/2511.16478/) (2025)
- [MUST-RAG: MUSical Text Question Answering with Retrieval Augmented Generation](https://mlsanity.com/paper/2507.23334/) (2025)
- [IncompeBench: A Permissively Licensed, Fine-Grained Benchmark for Music Information Retrieval](https://mlsanity.com/paper/2602.11941/) (2026)

## Abstract (arXiv)

Text-to-Music Retrieval, finding music based on a given natural language query, plays a pivotal role in content discovery within extensive music databases. To address this challenge, prior research has predominantly focused on a joint embedding of music audio and text, utilizing it to retrieve music tracks that exactly match descriptive queries related to musical attributes (i.e. genre, instrument) and contextual elements (i.e. mood, theme). However, users also articulate a need to explore music that shares similarities with their favorite tracks or artists, such as textitI need a similar track to Superstition by Stevie Wonder. To address these concerns, this paper proposes an improved Text-to-Music Retrieval model, denoted as TTMR++, which utilizes rich text descriptions generated with a finetuned large language model and metadata. To accomplish this, we obtained various types of seed text from several existing music tag and caption datasets and a knowledge graph dataset of artists and tracks. The experimental results show the effectiveness of TTMR++ in comparison to state-of-the-art music-text joint embedding models through a comprehensive evaluation involving various musical text queries.