Shypd.aiThe largest open index of AI toolsstable-audio-tools is a Content & Design tool that provides generative models for conditional audio generation. It offers training and inference code for audio generation models, installable via PyPI.
Trending
About
stable-audio-tools is a comprehensive library offering generative models for conditional audio generation. It provides both training and inference code, allowing users to develop and deploy their own audio generation models. The library is easily installable via PyPI and requires PyTorch 2.5 or later for advanced features like Flash Attention and Flex Attention. It supports various model types, including autoencoders and diffusion models for unconditional, conditional, and inpainting audio generation. A basic Gradio interface is included for testing trained models, and the system supports multi-GPU and multi-node training with PyTorch Lightning. Users can fine-tune pre-trained models or train from scratch using JSON configuration files for models and datasets.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending
