Audio Generative AI: Technologies and Implications

While the initial impact of generative AI has been in text or music applications, we are now seeing increasing interest in generative AI for audio applications, including for speech, music, and sound effects. In this talk, we will look at some of the technologies in this area, including AudioLDM latent diffusion models for audio. We will also cover some of the current ethical and legal issues raised by audio generative AI, such as the voice as personal data and copyright of creative content, and consider possible implications for the future of this area.