Please note that this website may not function properly on Safari and recommends using Google Chrome for the best experience. The website provides code to reproduce and evaluate Stable Audio, a model that can generate variable-length and long-form stereo music at 44.1kHz. The examples include different genres like Berlin techno, rave, and calming meditation music. The model can also generate stereo sound effects at 44.1kHz, such as door slams, sports cars passing by, and fireworks. The website compares the model’s performance with other state-of-the-art models and includes a qualitative study report. Additionally, there is a comparison of audio reconstructions using an autoencoder.
https://stability-ai.github.io/stable-audio-demo/