arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.12503cs.SDcs.AIcs.MMeess.ASeess.SP
收藏

AudioLDM:基于潜在扩散模型的文本到音频生成

AudioLDM: Text-to-Audio Generation with Latent Diffusion Models

University of Surrey(萨里大学);Imperial College London(帝国理工学院)

Haohe Liu, Zehua Chen, Yi Yuan, Xinhao Mei, Xubo Liu, Danilo Mandic, Wenwu Wang, Mark D. Plumbley

AI总结:

提出AudioLDM,基于CLAP潜在空间学习连续音频表示的文本到音频生成系统,在单GPU上训练即达到最优性能,并首次支持零样本文本引导音频操作。

摘要:

文本到音频(TTA)系统近期因能根据文本描述合成通用音频而受到关注。然而,以往的TTA研究生成质量有限且计算成本高。在本研究中,我们提出AudioLDM,这是一个构建在潜在空间上的TTA系统,旨在从对比语言-音频预训练(CLAP)潜在表示中学习连续音频表示。预训练的CLAP模型使我们能用音频嵌入训练LDM,同时在采样时提供文本嵌入作为条件。通过学习音频信号及其组成的潜在表示,而无需建模跨模态关系,AudioLDM在生成质量和计算效率上均具优势。在单GPU上使用AudioCaps训练,AudioLDM在客观和主观指标(如frechet distance)上均达到最先进的TTA性能。此外,AudioLDM是首个能以零样本方式实现多种文本引导音频操作(如风格迁移)的TTA系统。我们的实现和演示可在https://audioldm.github.io获取。

英文摘要:

Text-to-audio (TTA) system has recently gained attention for its ability to synthesize general audio based on text descriptions. However, previous studies in TTA have limited generation quality with high computational costs. In this study, we propose AudioLDM, a TTA system that is built on a latent space to learn the continuous audio representations from contrastive language-audio pretraining (CLAP) latents. The pretrained CLAP models enable us to train LDMs with audio embedding while providing text embedding as a condition during sampling. By learning the latent representations of audio signals and their compositions without modeling the cross-modal relationship, AudioLDM is advantageous in both generation quality and computational efficiency. Trained on AudioCaps with a single GPU, AudioLDM achieves state-of-the-art TTA performance measured by both objective and subjective metrics (e.g., frechet distance). Moreover, AudioLDM is the first TTA system that enables various text-guided audio manipulations (e.g., style transfer) in a zero-shot fashion. Our implementation and demos are available at https://audioldm.github.io.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑