IteraTTA: An interface for exploring both text prompts and audio priors in generating music with text-to-audio models

TL;DR

IteraTTA is an interactive interface that enables users to explore and refine both text prompts and audio priors in text-to-audio music generation, enhancing understanding and control over the generated music.

Contribution

This work introduces a novel interface, IteraTTA, that facilitates dual-sided exploration of text prompts and audio priors for improved music generation with text-to-audio models.

Findings

Users can effectively explore the impact of prompts and priors.

IteraTTA helps users understand the space of generated music.

Design considerations improve interaction with text-to-audio models.

Abstract

Recent text-to-audio generation techniques have the potential to allow novice users to freely generate music audio. Even if they do not have musical knowledge, such as about chord progressions and instruments, users can try various text prompts to generate audio. However, compared to the image domain, gaining a clear understanding of the space of possible music audios is difficult because users cannot listen to the variations of the generated audios simultaneously. We therefore facilitate users in exploring not only text prompts but also audio priors that constrain the text-to-audio music generation process. This dual-sided exploration enables users to discern the impact of different text prompts and audio priors on the generation results through iterative comparison of them. Our developed interface, IteraTTA, is specifically designed to aid users in refining text prompts and selecting favorable audio priors from the generated audios. With this, users can progressively reach their loosely-specified goals while understanding and exploring the space of possible results. Our implementation and discussions highlight design considerations that are specifically required for text-to-audio models and how interaction techniques can contribute to their effectiveness.