New Text-to-Audio Technology for Music Generation

The podcast discusses the advancements in the field of text-to-audio technology, specifically focusing on text-to-music capabilities. Several companies, including Google, Metas, and Cassette AI, are working on developing products that allow users to convert text into music. Stability AI has released its stable audio product, which uses generative AI techniques to deliver faster and higher quality music and sound effects. The model was trained using music and metadata from Audio Sparks, a music library with over 800,000 sounds. Stable audio offers both a free version and a pro subscription for commercial purposes.