
"Stability AI's Enhanced Stable Audio 2.0: Longer AI-Generated Songs, Better Quality?"
Stability AI's audio generation model, Stable Audio 2.0, now allows users to upload their own audio samples to create AI-generated songs up to three minutes long, but the results are still far from Grammy-worthy. The model, trained on data from AudioSparx, offers improved features for customization but still produces songs that sound soulless and weird, with some likening the vocals to whale sounds. Despite efforts to block copyrighted material, the model's ability to create songs that sound like actual songs remains a work in progress.



