AudiLDM

Generative soundscapes

AudioLDM is an AI system that uses latent diffusion models to generate realistic audio files such as sounds, music or speech from any text input. The model is open source and available on Github and Hugging Face.