FREEDOM251

Freedom251 › AI Sound Effect Generators

Sony Woosh

Free plan APISelf-hostedWeb

Visit Sony Woosh Alternatives

Sony Woosh's own site

What Sony Woosh is

Sony Woosh is Sony AI’s publicly released sound-effects foundation model, providing inference code and open model weights trained on public datasets. Woosh-Flow and Woosh-DFlow generate audio without a prompt or from text. Woosh-VFlow generates audio from a video sequence and can accept optional text prompts. The release includes Woosh-AE for audio encoding and decoding, Woosh-CLAP for text-audio alignment, text-to-audio and video-to-audio generators, and distilled generator versions. Installation supports CPU or CUDA, with inference test scripts for each model. Local Gradio demos for Woosh-Flow and Woosh-DFlow run in a browser on the same machine. A self-run inference API server is provided, currently for Woosh-DFlow only. GitHub releases provide pretrained weights and sample media; version 1.0.1 also includes SonicCLAP-AR for audio-text retrieval and SonicCLAP-MOS for alignment with human perceptual judgments. Most code is MIT-licensed, selected video-model code is Apache 2.0, and model weights are CC-BY-NC.

Who it suits

Woosh suits developers and researchers working on sound-effect generation, audio research, or model baselines. Its CPU/CUDA setup, local demos, and inference API support different workflows.

What it does well

  • Generates sound effects from text or video
  • Includes open weights and inference code
  • Supports CPU and CUDA installation
  • Offers local browser-based Gradio demos
  • Provides a self-run inference API

What to know first

  • API server currently supports Woosh-DFlow only
  • Model weights use a CC-BY-NC license
  • Commercial use is unclear

Verdict

Sony Woosh offers several sound-effect generation models with local demos, scripts, and open weights. The limited API model support and model-weight license should be reviewed before adoption.

Facts from Sony Woosh's own pages

Additional models
Release v1.0.1 provides SonicCLAP-AR for audio-text retrieval and SonicCLAP-MOS for alignment with human perceptual judgments. · github.com, 5 Oct 2026
API
A self-run API server is provided for inference, but currently supports Woosh-DFlow only. · github.com, 5 Oct 2026
Audience
The repository describes inference scripts, local demos, and an API server for running sound effect generation models. · github.com, 5 Oct 2026
Audio models
Woosh-AE provides audio encoding and decoding, while Woosh-CLAP provides text-audio alignment for diffusion conditioning. · github.com, 5 Oct 2026
Compute options
Installation instructions offer either CPU or CUDA support. · github.com, 5 Oct 2026
Downloads
Pretrained model weights and sample media are available through GitHub releases. · github.com, 5 Oct 2026
Inference options
The repository provides installation options with CPU or CUDA support and inference test scripts for each model. · github.com, 5 Oct 2026
Licensing
Most repository code is MIT-licensed, code for Woosh-VFlow and Woosh-DVFlow is Apache 2.0 licensed, and released model weights are CC-BY-NC licensed. · github.com, 5 Oct 2026
Local demos
The repository includes local Gradio demos for Woosh-Flow and Woosh-DFlow, accessed in a browser on the same machine. · github.com, 5 Oct 2026
Model components
The release includes an audio encoder/decoder, a text-audio alignment model, text-to-audio and video-to-audio generators, and distilled versions of the generators. · arxiv.org, 5 Oct 2026
Model weights
Open weights for models trained on public datasets are available from the repository’s releases. · github.com, 5 Oct 2026
Open weights
The repository provides inference code and open weights for models trained on public datasets. · github.com, 5 Oct 2026
Organization
The repository identifies Woosh as a public release of a sound effect foundation model by Sony AI. · github.com, 5 Oct 2026
Purpose
Woosh is Sony AI’s publicly released sound effects foundation model, with inference code and open model weights. · github.com, 5 Oct 2026
Support
The repository welcomes pull requests and asks contributors to open an issue before making major changes. · github.com, 5 Oct 2026
Text-to-audio
Woosh-Flow and Woosh-DFlow generate audio unconditionally or from a text prompt. · github.com, 5 Oct 2026
Use case
The models are optimized for sound effects and are presented as tools for audio research and building approaches or baselines. · arxiv.org, 5 Oct 2026
Video-to-audio
Woosh-VFlow generates audio from a video sequence and can take optional text prompts. · github.com, 5 Oct 2026

Where Sony Woosh stands on Freedom251

Look at these too