README.md

---
title: "Woosh-DFlow (Text-to-Audio SFX)"
colorFrom: indigo
colorTo: gray
sdk: gradio
sdk_version: 5.28.0
app_file: app.py
pinned: false
license: "cc-by-nc-4.0"
---

# Woosh-DFlow (Text-to-Audio SFX)

Generate a ~5s, 48kHz sound effect from a text prompt using Sony AI's Woosh-DFlow, the distilled (4-step) text-to-audio model from the Woosh sound-effect foundation model family. This is a thin HARP frontend that proxies to a Woosh-DFlow backend Space over its /generate API; the heavy model (Python 3.12, torch 2.8, Gradio 6) runs there, unmodified. Open weights are CC-BY-NC 4.0 (non-commercial).

- Inputs: textbox, slider, number
- Outputs: audio

Generated by the HARP model agent from a recipe.