Text2midi: Generating Symbolic Music from Captions

TL;DR

Text2midi is an innovative end-to-end system that uses large language models to generate MIDI music files from textual descriptions, enabling intuitive and controllable music creation.

Contribution

It introduces a novel approach combining LLMs and autoregressive transformers to generate symbolic music from text, streamlining the music composition process.

Findings

High-quality MIDI generation controllable by text prompts

Effective use of LLMs for symbolic music synthesis

Positive results from automated and human evaluations

Abstract

This paper introduces text2midi, an end-to-end model to generate MIDI files from textual descriptions. Leveraging the growing popularity of multimodal generative approaches, text2midi capitalizes on the extensive availability of textual data and the success of large language models (LLMs). Our end-to-end system harnesses the power of LLMs to generate symbolic music in the form of MIDI files. Specifically, we utilize a pretrained LLM encoder to process captions, which then condition an autoregressive transformer decoder to produce MIDI sequences that accurately reflect the provided descriptions. This intuitive and user-friendly method significantly streamlines the music creation process by allowing users to generate music pieces using text prompts. We conduct comprehensive empirical evaluations, incorporating both automated and human studies, that show our model generates MIDI files of high quality that are indeed controllable by text captions that may include music theory terms such as chords, keys, and tempo. We release the code and music samples on our demo page (https://github.com/AMAAI-Lab/Text2midi) for users to interact with text2midi.