• jonjuan
    link
    fedilink
    English
    arrow-up
    1
    ·
    1 month ago

    They aren’t using a LLM, this is a speech to text model like Whisper.