Woosh: A Sound Effects Foundation Model
This provides a foundational tool for the audio research community to build novel approaches and establish baselines, but it is incremental as it builds upon existing open models with optimizations for sound effects.
The paper tackles the problem of generating sound effects by introducing Woosh, a publicly released foundation model from Sony AI, which includes audio encoder/decoder, text-audio alignment, and generative models for text-to-audio and video-to-audio tasks, with evaluation showing competitive or better performance compared to existing open models like StableAudio-Open and TangoFlux.
The audio research community depends on open generative models as foundational tools for building novel approaches and establishing baselines. In this report, we present Woosh, Sony AI's publicly released sound effect foundation model, detailing its architecture, training process, and an evaluation against other popular open models. Being optimized for sound effects, we provide (1) a high-quality audio encoder/decoder model and (2) a text-audio alignment model for conditioning, together with (3) text-to-audio and (4) video-to-audio generative models. Distilled text-to-audio and video-to-audio models are also included in the release, allowing for low-resource operation and fast inference. Our evaluation on both public and private data shows competitive or better performance for each module when compared to existing open alternatives like StableAudio-Open and TangoFlux. Inference code and model weights are available at https://github.com/SonyResearch/Woosh. Demo samples can be found at https://sonyresearch.github.io/Woosh/.