Audio ControlNet for Fine-Grained Audio Generation and Editing
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917248037289984 |
|---|---|
| author | Zhu, Haina Xiao, Yao Li, Xiquan Ma, Ziyang Yu, Jianwei Zhang, Bowen Yang, Mingqi Chen, Xie |
| author_facet | Zhu, Haina Xiao, Yao Li, Xiquan Ma, Ziyang Yu, Jianwei Zhang, Bowen Yang, Mingqi Chen, Xie |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_04680 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Audio ControlNet for Fine-Grained Audio Generation and Editing Zhu, Haina Xiao, Yao Li, Xiquan Ma, Ziyang Yu, Jianwei Zhang, Bowen Yang, Mingqi Chen, Xie Sound Artificial Intelligence Computation and Language Multimedia 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. |
| title | Audio ControlNet for Fine-Grained Audio Generation and Editing |
| topic | Sound Artificial Intelligence Computation and Language Multimedia |
| url | https://arxiv.org/abs/2602.04680 |