Meta’s New Open-Source SAM Audio AI Model Can Isolate Sounds From Audio Mixtures
Meta has launched SAM Audio, a new open-source AI model that can identify and isolate specific sounds from complex audio. Part of the Segment Anything Model family, it supports text prompts, visual cues and time-based inputs to simplify audio editing. Meta says the tool can automate tasks that usually require specialised software, and it is available for both research and commercial use.
