AudioLDM 2: Learning Holistic Audio Generation with Self-supervised Pretraining
read the original abstract
Although audio generation shares commonalities across different types of audio, such as speech, music, and sound effects, designing models for each type requires careful consideration of specific objectives and biases that can significantly differ from those of other types. To bring us closer to a unified perspective of audio generation, this paper proposes a framework that utilizes the same learning method for speech, music, and sound effect generation. Our framework introduces a general representation of audio, called "language of audio" (LOA). Any audio can be translated into LOA based on AudioMAE, a self-supervised pre-trained representation learning model. In the generation process, we translate any modalities into LOA by using a GPT-2 model, and we perform self-supervised audio generation learning with a latent diffusion model conditioned on LOA. The proposed framework naturally brings advantages such as in-context learning abilities and reusable self-supervised pretrained AudioMAE and latent diffusion models. Experiments on the major benchmarks of text-to-audio, text-to-music, and text-to-speech demonstrate state-of-the-art or competitive performance against previous approaches. Our code, pretrained model, and demo are available at https://audioldm.github.io/audioldm2.
This paper has not been read by Pith yet.
Forward citations
Cited by 5 Pith papers
-
LeVo 2: Stable and Melodious Song Generation via Hierarchical Representation Modeling and Progressive Post-Training
LeVo 2 presents a hierarchical LLM-Diffusion model with progressive post-training stages to generate full-length songs that balance semantic planning, track-specific acoustics, and musicality.
-
Inside the Latent Flow: Causal Deciphering of Attention Dynamics in Audio Separation Foundation Models
Causal probing of attention in audio separation transformers identifies dual pathways and asynchronous convergence, enabling a training-free Layer-Selective Attention Caching method that reduces self-attention computa...
-
UNISON: A Unified Sound Generation and Editing Framework via Deep LLM Fusion
UNISON introduces a unified latent diffusion framework with layer-wise LLM fusion and channel-mask task encoding for multiple speech and sound generation and editing tasks.
-
Woosh: A Sound Effects Foundation Model
Woosh is a new publicly released foundation model optimized for high-quality sound effect generation from text or video, showing competitive or better results than open alternatives like Stable Audio Open.
-
Training-Free Multi-User Generative Semantic Communications via Null-Space Diffusion Sampling
Introduces a null-space diffusion sampling method for training-free multi-user generative semantic communications in OFDMA systems.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.