flâneur

Woosh: A Sound Effects Foundation Model

arxiv.org · 10,874 words · saved by 1 readers

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 h

Woosh: A Sound Effects Foundation Model Gaëtan Hadjeres 1 Marc Ferras 1 Khaled Koutini 1 Benno Weck 1 Work done during an internship at Sony AI. Alexandre Bittar 1∗ Thomas Hummel 1∗ Zineb Lahrici 1 Hakim Missoum 1 Joan Serrà 1 and Yuki Mitsufuji 1,2 Abstract 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

saved by

related reading