Evaluate model predictions against ground truth data
Stable audio open model from Synthio paper.
Generate text based on audio input and questions
Describe audio with questions
Create images in seconds. No sign-up, no paywall, no setup.