---
title: Element-16 Audio Guidance
emoji: 🎵
colorFrom: purple
colorTo: pink
sdk: gradio
sdk_version: 6.8.0
python_version: '3.12'
app_file: app.py
pinned: false
short_description: Audio-guided video generation with Sulphur (LTX-2.4 base)
---

# Element-16 Audio Guidance

Audio-conditioned video generation using Sulphur (LTX-2.4 base model) with the native `A2VidPipelineTwoStage` pipeline.

## Features
- Audio as input conditioning (lipsync / audio-reactive motion)
- Optional first frame image conditioning
- Up to 20 seconds at 512×896 (portrait) by default
- Two-stage pipeline: full inference stage 1 + distilled LoRA stage 2
- Adjustable audio guidance scale and CFG