STABILITY AI TEXT-TO-AUDIO MODEL

Summary from the TLDR AI newsletter

Stability AI has open-sourced Stable Audio Open Small, a 341M parameter text-to-audio model optimized to run on Arm CPUs. It can generate 11-second audio clips on smartphones in under 8 seconds.