A Streamlit application that converts audio files into videos with synchronized images.
- Audio transcription using Whisper-large-v3
- Image scraping from web sources
- Video generation with multiple format options (16:9, 9:16, 1:1)
- Configurable segment duration
- Downloadable video output
- Push this repository to GitHub
- Connect to Streamlit Cloud
- Set the app file to
app.py - Add
setup.shas the "Run" command in Advanced Settings - Deploy!
See requirements.txt for full list of dependencies.
- Upload an audio file (max 50 minutes)
- Adjust settings (images per keyword, segment duration, video format, quality)
- Click "Generate Video"
- Download the resulting video