Audio ControlNet for Fine-Grained Audio Generation and Editing

TL;DR

This paper introduces a controllable text-to-audio generation framework using ControlNet and adapters, enabling precise control over audio attributes and editing with minimal additional parameters, advancing the state-of-the-art.

Contribution

It proposes T2A-ControlNet and T2A-Adapter models for controllable audio synthesis and editing, with the latter being more efficient and effective, and achieves state-of-the-art results.

Findings

T2A-Adapter achieves state-of-the-art F1 scores on AudioSet-Strong.

The models enable fine-grained control over loudness, pitch, and sound events.

The framework supports audio editing tasks like event removal and insertion.

Abstract

We study the fine-grained text-to-audio (T2A) generation task. While recent models can synthesize high-quality audio from text descriptions, they often lack precise control over attributes such as loudness, pitch, and sound events. Unlike prior approaches that retrain models for specific control types, we propose to train ControlNet models on top of pre-trained T2A backbones to achieve controllable generation over loudness, pitch, and event roll. We introduce two designs, T2A-ControlNet and T2A-Adapter, and show that the T2A-Adapter model offers a more efficient structure with strong control ability. With only 38M additional parameters, T2A-Adapter achieves state-of-the-art performance on the AudioSet-Strong in both event-level and segment-level F1 scores. We further extend this framework to audio editing, proposing T2A-Editor for removing and inserting audio events at time locations specified by instructions. Models, code, dataset pipelines, and benchmarks will be released to support future research on controllable audio generation and editing.