Towards Diverse and Efficient Audio Captioning via Diffusion Models

TL;DR

This paper presents Diffusion-based Audio Captioning (DAC), a novel non-autoregressive diffusion model that achieves state-of-the-art caption quality while significantly improving diversity and generation speed in audio captioning tasks.

Contribution

The paper introduces DAC, a diffusion model for audio captioning that enhances diversity, speed, and quality, and demonstrates its potential for unified multimodal generation.

Findings

DAC achieves SOTA caption quality.

DAC significantly improves diversity and speed.

Diffusion models can unify audio, visual, and text generation.

Abstract

We introduce Diffusion-based Audio Captioning (DAC), a non-autoregressive diffusion model tailored for diverse and efficient audio captioning. Although existing captioning models relying on language backbones have achieved remarkable success in various captioning tasks, their insufficient performance in terms of generation speed and diversity impede progress in audio understanding and multimedia applications. Our diffusion-based framework offers unique advantages stemming from its inherent stochasticity and holistic context modeling in captioning. Through rigorous evaluation, we demonstrate that DAC not only achieves SOTA performance levels compared to existing benchmarks in the caption quality, but also significantly outperforms them in terms of generation speed and diversity. The success of DAC illustrates that text generation can also be seamlessly integrated with audio and visual generation tasks using a diffusion backbone, paving the way for a unified, audio-related generative model across different modalities.