Google’s AI can play a song snippet and then continue to do so
After being given a few seconds’ audio, a new AI system is able to create music and speech that sounds natural.
AudioLM is a Google-developed tool that generates audio in the style of a prompt. This includes complex sounds such as piano music or people talking, and it’s almost impossible to tell the difference between the original recording and the generated audio. This technique could be used in the future to automatically generate music for videos.