A5S Async V1 is a speech recognition model that generates speaker-attributed transcripts from recorded audio files. It assigns labels to speakers, adds timestamps, and supports export to subtitle formats through an asynchronous API.